diff options
author | Linus Torvalds <torvalds@linux-foundation.org> | 2015-06-24 20:47:21 -0700 |
---|---|---|
committer | Linus Torvalds <torvalds@linux-foundation.org> | 2015-06-24 20:47:21 -0700 |
commit | aefbef10e3ae6e2c6e3c54f906f10b34c73a2c66 (patch) | |
tree | ef967a568ff5e7bb52d1d3d0c61e701ad4f31c21 /mm | |
parent | 266da6f14232638b9caafb7facf2a7333895dd05 (diff) | |
parent | 8a8c35fadfaf55629a37ef1a8ead1b8fb32581d2 (diff) |
Merge branch 'akpm' (patches from Andrew)
Merge first patchbomb from Andrew Morton:
- a few misc things
- ocfs2 udpates
- kernel/watchdog.c feature work (took ages to get right)
- most of MM. A few tricky bits are held up and probably won't make 4.2.
* emailed patches from Andrew Morton <akpm@linux-foundation.org>: (91 commits)
mm: kmemleak_alloc_percpu() should follow the gfp from per_alloc()
mm, thp: respect MPOL_PREFERRED policy with non-local node
tmpfs: truncate prealloc blocks past i_size
mm/memory hotplug: print the last vmemmap region at the end of hot add memory
mm/mmap.c: optimization of do_mmap_pgoff function
mm: kmemleak: optimise kmemleak_lock acquiring during kmemleak_scan
mm: kmemleak: avoid deadlock on the kmemleak object insertion error path
mm: kmemleak: do not acquire scan_mutex in kmemleak_do_cleanup()
mm: kmemleak: fix delete_object_*() race when called on the same memory block
mm: kmemleak: allow safe memory scanning during kmemleak disabling
memcg: convert mem_cgroup->under_oom from atomic_t to int
memcg: remove unused mem_cgroup->oom_wakeups
frontswap: allow multiple backends
x86, mirror: x86 enabling - find mirrored memory ranges
mm/memblock: allocate boot time data structures from mirrored memory
mm/memblock: add extra "flags" to memblock to allow selection of memory based on attribute
mm: do not ignore mapping_gfp_mask in page cache allocation paths
mm/cma.c: fix typos in comments
mm/oom_kill.c: print points as unsigned int
mm/hugetlb: handle races in alloc_huge_page and hugetlb_reserve_pages
...
Diffstat (limited to 'mm')
-rw-r--r-- | mm/Kconfig | 1 | ||||
-rw-r--r-- | mm/cma.c | 10 | ||||
-rw-r--r-- | mm/filemap.c | 23 | ||||
-rw-r--r-- | mm/frontswap.c | 215 | ||||
-rw-r--r-- | mm/huge_memory.c | 22 | ||||
-rw-r--r-- | mm/hugetlb.c | 187 | ||||
-rw-r--r-- | mm/hwpoison-inject.c | 4 | ||||
-rw-r--r-- | mm/kmemleak.c | 168 | ||||
-rw-r--r-- | mm/memblock.c | 123 | ||||
-rw-r--r-- | mm/memcontrol.c | 59 | ||||
-rw-r--r-- | mm/memory-failure.c | 351 | ||||
-rw-r--r-- | mm/memory.c | 10 | ||||
-rw-r--r-- | mm/memory_hotplug.c | 1 | ||||
-rw-r--r-- | mm/mempolicy.c | 38 | ||||
-rw-r--r-- | mm/memtest.c | 3 | ||||
-rw-r--r-- | mm/migrate.c | 11 | ||||
-rw-r--r-- | mm/mmap.c | 6 | ||||
-rw-r--r-- | mm/mprotect.c | 11 | ||||
-rw-r--r-- | mm/mremap.c | 17 | ||||
-rw-r--r-- | mm/nobootmem.c | 14 | ||||
-rw-r--r-- | mm/nommu.c | 112 | ||||
-rw-r--r-- | mm/oom_kill.c | 158 | ||||
-rw-r--r-- | mm/page_alloc.c | 177 | ||||
-rw-r--r-- | mm/percpu.c | 2 | ||||
-rw-r--r-- | mm/pgtable-generic.c | 29 | ||||
-rw-r--r-- | mm/rmap.c | 9 | ||||
-rw-r--r-- | mm/shmem.c | 2 | ||||
-rw-r--r-- | mm/slab.c | 1 | ||||
-rw-r--r-- | mm/slab.h | 1 | ||||
-rw-r--r-- | mm/slab_common.c | 98 | ||||
-rw-r--r-- | mm/slub.c | 1 | ||||
-rw-r--r-- | mm/swap.c | 1 | ||||
-rw-r--r-- | mm/vmscan.c | 15 |
33 files changed, 1009 insertions, 871 deletions
diff --git a/mm/Kconfig b/mm/Kconfig index 390214da4546..c180af880ed5 100644 --- a/mm/Kconfig +++ b/mm/Kconfig @@ -368,6 +368,7 @@ config MEMORY_FAILURE depends on ARCH_SUPPORTS_MEMORY_FAILURE bool "Enable recovery from hardware memory errors" select MEMORY_ISOLATION + select RAS help Enables code to recover from some memory failures on systems with MCA recovery. This allows a system to continue running @@ -182,7 +182,7 @@ int __init cma_init_reserved_mem(phys_addr_t base, phys_addr_t size, if (!size || !memblock_is_region_reserved(base, size)) return -EINVAL; - /* ensure minimal alignment requied by mm core */ + /* ensure minimal alignment required by mm core */ alignment = PAGE_SIZE << max(MAX_ORDER - 1, pageblock_order); /* alignment should be aligned with order_per_bit */ @@ -238,7 +238,7 @@ int __init cma_declare_contiguous(phys_addr_t base, /* * high_memory isn't direct mapped memory so retrieving its physical * address isn't appropriate. But it would be useful to check the - * physical address of the highmem boundary so it's justfiable to get + * physical address of the highmem boundary so it's justifiable to get * the physical address from it. On x86 there is a validation check for * this case, so the following workaround is needed to avoid it. */ @@ -316,13 +316,15 @@ int __init cma_declare_contiguous(phys_addr_t base, */ if (base < highmem_start && limit > highmem_start) { addr = memblock_alloc_range(size, alignment, - highmem_start, limit); + highmem_start, limit, + MEMBLOCK_NONE); limit = highmem_start; } if (!addr) { addr = memblock_alloc_range(size, alignment, base, - limit); + limit, + MEMBLOCK_NONE); if (!addr) { ret = -ENOMEM; goto err; diff --git a/mm/filemap.c b/mm/filemap.c index 6bf5e42d560a..8d17ceea8dbe 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -196,7 +196,9 @@ void __delete_from_page_cache(struct page *page, void *shadow) page->mapping = NULL; /* Leave page->index set: truncation lookup relies upon it */ - __dec_zone_page_state(page, NR_FILE_PAGES); + /* hugetlb pages do not participate in page cache accounting. */ + if (!PageHuge(page)) + __dec_zone_page_state(page, NR_FILE_PAGES); if (PageSwapBacked(page)) __dec_zone_page_state(page, NR_SHMEM); BUG_ON(page_mapped(page)); @@ -483,7 +485,12 @@ int replace_page_cache_page(struct page *old, struct page *new, gfp_t gfp_mask) error = radix_tree_insert(&mapping->page_tree, offset, new); BUG_ON(error); mapping->nrpages++; - __inc_zone_page_state(new, NR_FILE_PAGES); + + /* + * hugetlb pages do not participate in page cache accounting. + */ + if (!PageHuge(new)) + __inc_zone_page_state(new, NR_FILE_PAGES); if (PageSwapBacked(new)) __inc_zone_page_state(new, NR_SHMEM); spin_unlock_irq(&mapping->tree_lock); @@ -575,7 +582,10 @@ static int __add_to_page_cache_locked(struct page *page, radix_tree_preload_end(); if (unlikely(error)) goto err_insert; - __inc_zone_page_state(page, NR_FILE_PAGES); + + /* hugetlb pages do not participate in page cache accounting. */ + if (!huge) + __inc_zone_page_state(page, NR_FILE_PAGES); spin_unlock_irq(&mapping->tree_lock); if (!huge) mem_cgroup_commit_charge(page, memcg, false); @@ -1654,8 +1664,8 @@ no_cached_page: error = -ENOMEM; goto out; } - error = add_to_page_cache_lru(page, mapping, - index, GFP_KERNEL); + error = add_to_page_cache_lru(page, mapping, index, + GFP_KERNEL & mapping_gfp_mask(mapping)); if (error) { page_cache_release(page); if (error == -EEXIST) { @@ -1756,7 +1766,8 @@ static int page_cache_read(struct file *file, pgoff_t offset) if (!page) return -ENOMEM; - ret = add_to_page_cache_lru(page, mapping, offset, GFP_KERNEL); + ret = add_to_page_cache_lru(page, mapping, offset, + GFP_KERNEL & mapping_gfp_mask(mapping)); if (ret == 0) ret = mapping->a_ops->readpage(file, page); else if (ret == -EEXIST) diff --git a/mm/frontswap.c b/mm/frontswap.c index 8d82809eb085..27a9924caf61 100644 --- a/mm/frontswap.c +++ b/mm/frontswap.c @@ -21,11 +21,16 @@ #include <linux/swapfile.h> /* - * frontswap_ops is set by frontswap_register_ops to contain the pointers - * to the frontswap "backend" implementation functions. + * frontswap_ops are added by frontswap_register_ops, and provide the + * frontswap "backend" implementation functions. Multiple implementations + * may be registered, but implementations can never deregister. This + * is a simple singly-linked list of all registered implementations. */ static struct frontswap_ops *frontswap_ops __read_mostly; +#define for_each_frontswap_ops(ops) \ + for ((ops) = frontswap_ops; (ops); (ops) = (ops)->next) + /* * If enabled, frontswap_store will return failure even on success. As * a result, the swap subsystem will always write the page to swap, in @@ -79,15 +84,6 @@ static inline void inc_frontswap_invalidates(void) { } * on all frontswap functions to not call the backend until the backend * has registered. * - * Specifically when no backend is registered (nobody called - * frontswap_register_ops) all calls to frontswap_init (which is done via - * swapon -> enable_swap_info -> frontswap_init) are registered and remembered - * (via the setting of need_init bitmap) but fail to create tmem_pools. When a - * backend registers with frontswap at some later point the previous - * calls to frontswap_init are executed (by iterating over the need_init - * bitmap) to create tmem_pools and set the respective poolids. All of that is - * guarded by us using atomic bit operations on the 'need_init' bitmap. - * * This would not guards us against the user deciding to call swapoff right as * we are calling the backend to initialize (so swapon is in action). * Fortunatly for us, the swapon_mutex has been taked by the callee so we are @@ -106,37 +102,64 @@ static inline void inc_frontswap_invalidates(void) { } * * Obviously the opposite (unloading the backend) must be done after all * the frontswap_[store|load|invalidate_area|invalidate_page] start - * ignorning or failing the requests - at which point frontswap_ops - * would have to be made in some fashion atomic. + * ignoring or failing the requests. However, there is currently no way + * to unload a backend once it is registered. */ -static DECLARE_BITMAP(need_init, MAX_SWAPFILES); /* - * Register operations for frontswap, returning previous thus allowing - * detection of multiple backends and possible nesting. + * Register operations for frontswap */ -struct frontswap_ops *frontswap_register_ops(struct frontswap_ops *ops) +void frontswap_register_ops(struct frontswap_ops *ops) { - struct frontswap_ops *old = frontswap_ops; - int i; - - for (i = 0; i < MAX_SWAPFILES; i++) { - if (test_and_clear_bit(i, need_init)) { - struct swap_info_struct *sis = swap_info[i]; - /* __frontswap_init _should_ have set it! */ - if (!sis->frontswap_map) - return ERR_PTR(-EINVAL); - ops->init(i); - } + DECLARE_BITMAP(a, MAX_SWAPFILES); + DECLARE_BITMAP(b, MAX_SWAPFILES); + struct swap_info_struct *si; + unsigned int i; + + bitmap_zero(a, MAX_SWAPFILES); + bitmap_zero(b, MAX_SWAPFILES); + + spin_lock(&swap_lock); + plist_for_each_entry(si, &swap_active_head, list) { + if (!WARN_ON(!si->frontswap_map)) + set_bit(si->type, a); } + spin_unlock(&swap_lock); + + /* the new ops needs to know the currently active swap devices */ + for_each_set_bit(i, a, MAX_SWAPFILES) + ops->init(i); + /* - * We MUST have frontswap_ops set _after_ the frontswap_init's - * have been called. Otherwise __frontswap_store might fail. Hence - * the barrier to make sure compiler does not re-order us. + * Setting frontswap_ops must happen after the ops->init() calls + * above; cmpxchg implies smp_mb() which will ensure the init is + * complete at this point. */ - barrier(); - frontswap_ops = ops; - return old; + do { + ops->next = frontswap_ops; + } while (cmpxchg(&frontswap_ops, ops->next, ops) != ops->next); + + spin_lock(&swap_lock); + plist_for_each_entry(si, &swap_active_head, list) { + if (si->frontswap_map) + set_bit(si->type, b); + } + spin_unlock(&swap_lock); + + /* + * On the very unlikely chance that a swap device was added or + * removed between setting the "a" list bits and the ops init + * calls, we re-check and do init or invalidate for any changed + * bits. + */ + if (unlikely(!bitmap_equal(a, b, MAX_SWAPFILES))) { + for (i = 0; i < MAX_SWAPFILES; i++) { + if (!test_bit(i, a) && test_bit(i, b)) + ops->init(i); + else if (test_bit(i, a) && !test_bit(i, b)) + ops->invalidate_area(i); + } + } } EXPORT_SYMBOL(frontswap_register_ops); @@ -164,6 +187,7 @@ EXPORT_SYMBOL(frontswap_tmem_exclusive_gets); void __frontswap_init(unsigned type, unsigned long *map) { struct swap_info_struct *sis = swap_info[type]; + struct frontswap_ops *ops; BUG_ON(sis == NULL); @@ -179,28 +203,30 @@ void __frontswap_init(unsigned type, unsigned long *map) * p->frontswap set to something valid to work properly. */ frontswap_map_set(sis, map); - if (frontswap_ops) - frontswap_ops->init(type); - else { - BUG_ON(type >= MAX_SWAPFILES); - set_bit(type, need_init); - } + + for_each_frontswap_ops(ops) + ops->init(type); } EXPORT_SYMBOL(__frontswap_init); bool __frontswap_test(struct swap_info_struct *sis, pgoff_t offset) { - bool ret = false; - - if (frontswap_ops && sis->frontswap_map) - ret = test_bit(offset, sis->frontswap_map); - return ret; + if (sis->frontswap_map) + return test_bit(offset, sis->frontswap_map); + return false; } EXPORT_SYMBOL(__frontswap_test); +static inline void __frontswap_set(struct swap_info_struct *sis, + pgoff_t offset) +{ + set_bit(offset, sis->frontswap_map); + atomic_inc(&sis->frontswap_pages); +} + static inline void __frontswap_clear(struct swap_info_struct *sis, - pgoff_t offset) + pgoff_t offset) { clear_bit(offset, sis->frontswap_map); atomic_dec(&sis->frontswap_pages); @@ -215,39 +241,46 @@ static inline void __frontswap_clear(struct swap_info_struct *sis, */ int __frontswap_store(struct page *page) { - int ret = -1, dup = 0; + int ret = -1; swp_entry_t entry = { .val = page_private(page), }; int type = swp_type(entry); struct swap_info_struct *sis = swap_info[type]; pgoff_t offset = swp_offset(entry); + struct frontswap_ops *ops; /* * Return if no backend registed. * Don't need to inc frontswap_failed_stores here. */ if (!frontswap_ops) - return ret; + return -1; BUG_ON(!PageLocked(page)); BUG_ON(sis == NULL); - if (__frontswap_test(sis, offset)) - dup = 1; - ret = frontswap_ops->store(type, offset, page); + + /* + * If a dup, we must remove the old page first; we can't leave the + * old page no matter if the store of the new page succeeds or fails, + * and we can't rely on the new page replacing the old page as we may + * not store to the same implementation that contains the old page. + */ + if (__frontswap_test(sis, offset)) { + __frontswap_clear(sis, offset); + for_each_frontswap_ops(ops) + ops->invalidate_page(type, offset); + } + + /* Try to store in each implementation, until one succeeds. */ + for_each_frontswap_ops(ops) { + ret = ops->store(type, offset, page); + if (!ret) /* successful store */ + break; + } if (ret == 0) { - set_bit(offset, sis->frontswap_map); + __frontswap_set(sis, offset); inc_frontswap_succ_stores(); - if (!dup) - atomic_inc(&sis->frontswap_pages); } else { - /* - failed dup always results in automatic invalidate of - the (older) page from frontswap - */ inc_frontswap_failed_stores(); - if (dup) { - __frontswap_clear(sis, offset); - frontswap_ops->invalidate_page(type, offset); - } } if (frontswap_writethrough_enabled) /* report failure so swap also writes to swap device */ @@ -268,14 +301,22 @@ int __frontswap_load(struct page *page) int type = swp_type(entry); struct swap_info_struct *sis = swap_info[type]; pgoff_t offset = swp_offset(entry); + struct frontswap_ops *ops; + + if (!frontswap_ops) + return -1; BUG_ON(!PageLocked(page)); BUG_ON(sis == NULL); - /* - * __frontswap_test() will check whether there is backend registered - */ - if (__frontswap_test(sis, offset)) - ret = frontswap_ops->load(type, offset, page); + if (!__frontswap_test(sis, offset)) + return -1; + + /* Try loading from each implementation, until one succeeds. */ + for_each_frontswap_ops(ops) { + ret = ops->load(type, offset, page); + if (!ret) /* successful load */ + break; + } if (ret == 0) { inc_frontswap_loads(); if (frontswap_tmem_exclusive_gets_enabled) { @@ -294,16 +335,19 @@ EXPORT_SYMBOL(__frontswap_load); void __frontswap_invalidate_page(unsigned type, pgoff_t offset) { struct swap_info_struct *sis = swap_info[type]; + struct frontswap_ops *ops; + + if (!frontswap_ops) + return; BUG_ON(sis == NULL); - /* - * __frontswap_test() will check whether there is backend registered - */ - if (__frontswap_test(sis, offset)) { - frontswap_ops->invalidate_page(type, offset); - __frontswap_clear(sis, offset); - inc_frontswap_invalidates(); - } + if (!__frontswap_test(sis, offset)) + return; + + for_each_frontswap_ops(ops) + ops->invalidate_page(type, offset); + __frontswap_clear(sis, offset); + inc_frontswap_invalidates(); } EXPORT_SYMBOL(__frontswap_invalidate_page); @@ -314,16 +358,19 @@ EXPORT_SYMBOL(__frontswap_invalidate_page); void __frontswap_invalidate_area(unsigned type) { struct swap_info_struct *sis = swap_info[type]; + struct frontswap_ops *ops; - if (frontswap_ops) { - BUG_ON(sis == NULL); - if (sis->frontswap_map == NULL) - return; - frontswap_ops->invalidate_area(type); - atomic_set(&sis->frontswap_pages, 0); - bitmap_zero(sis->frontswap_map, sis->max); - } - clear_bit(type, need_init); + if (!frontswap_ops) + return; + + BUG_ON(sis == NULL); + if (sis->frontswap_map == NULL) + return; + + for_each_frontswap_ops(ops) + ops->invalidate_area(type); + atomic_set(&sis->frontswap_pages, 0); + bitmap_zero(sis->frontswap_map, sis->max); } EXPORT_SYMBOL(__frontswap_invalidate_area); diff --git a/mm/huge_memory.c b/mm/huge_memory.c index 078832cf3636..c107094f79ba 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -1031,7 +1031,7 @@ static int do_huge_pmd_wp_page_fallback(struct mm_struct *mm, goto out_free_pages; VM_BUG_ON_PAGE(!PageHead(page), page); - pmdp_clear_flush_notify(vma, haddr, pmd); + pmdp_huge_clear_flush_notify(vma, haddr, pmd); /* leave pmd empty until pte is filled */ pgtable = pgtable_trans_huge_withdraw(mm, pmd); @@ -1174,7 +1174,7 @@ alloc: pmd_t entry; entry = mk_huge_pmd(new_page, vma->vm_page_prot); entry = maybe_pmd_mkwrite(pmd_mkdirty(entry), vma); - pmdp_clear_flush_notify(vma, haddr, pmd); + pmdp_huge_clear_flush_notify(vma, haddr, pmd); page_add_new_anon_rmap(new_page, vma, haddr); mem_cgroup_commit_charge(new_page, memcg, false); lru_cache_add_active_or_unevictable(new_page, vma); @@ -1396,12 +1396,12 @@ int zap_huge_pmd(struct mmu_gather *tlb, struct vm_area_struct *vma, pmd_t orig_pmd; /* * For architectures like ppc64 we look at deposited pgtable - * when calling pmdp_get_and_clear. So do the + * when calling pmdp_huge_get_and_clear. So do the * pgtable_trans_huge_withdraw after finishing pmdp related * operations. */ - orig_pmd = pmdp_get_and_clear_full(tlb->mm, addr, pmd, - tlb->fullmm); + orig_pmd = pmdp_huge_get_and_clear_full(tlb->mm, addr, pmd, + tlb->fullmm); tlb_remove_pmd_tlb_entry(tlb, pmd, addr); pgtable = pgtable_trans_huge_withdraw(tlb->mm, pmd); if (is_huge_zero_pmd(orig_pmd)) { @@ -1459,7 +1459,7 @@ int move_huge_pmd(struct vm_area_struct *vma, struct vm_area_struct *new_vma, new_ptl = pmd_lockptr(mm, new_pmd); if (new_ptl != old_ptl) spin_lock_nested(new_ptl, SINGLE_DEPTH_NESTING); - pmd = pmdp_get_and_clear(mm, old_addr, old_pmd); + pmd = pmdp_huge_get_and_clear(mm, old_addr, old_pmd); VM_BUG_ON(!pmd_none(*new_pmd)); if (pmd_move_must_withdraw(new_ptl, old_ptl)) { @@ -1505,7 +1505,7 @@ int change_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd, } if (!prot_numa || !pmd_protnone(*pmd)) { - entry = pmdp_get_and_clear_notify(mm, addr, pmd); + entry = pmdp_huge_get_and_clear_notify(mm, addr, pmd); entry = pmd_modify(entry, newprot); if (preserve_write) entry = pmd_mkwrite(entry); @@ -2499,7 +2499,7 @@ static void collapse_huge_page(struct mm_struct *mm, * huge and small TLB entries for the same virtual address * to avoid the risk of CPU bugs in that area. */ - _pmd = pmdp_clear_flush(vma, address, pmd); + _pmd = pmdp_collapse_flush(vma, address, pmd); spin_unlock(pmd_ptl); mmu_notifier_invalidate_range_end(mm, mmun_start, mmun_end); @@ -2799,7 +2799,7 @@ static void khugepaged_do_scan(void) cond_resched(); - if (unlikely(kthread_should_stop() || freezing(current))) + if (unlikely(kthread_should_stop() || try_to_freeze())) break; spin_lock(&khugepaged_mm_lock); @@ -2820,8 +2820,6 @@ static void khugepaged_do_scan(void) static void khugepaged_wait_work(void) { - try_to_freeze(); - if (khugepaged_has_work()) { if (!khugepaged_scan_sleep_millisecs) return; @@ -2865,7 +2863,7 @@ static void __split_huge_zero_page_pmd(struct vm_area_struct *vma, pmd_t _pmd; int i; - pmdp_clear_flush_notify(vma, haddr, pmd); + pmdp_huge_clear_flush_notify(vma, haddr, pmd); /* leave pmd empty until pte is filled */ pgtable = pgtable_trans_huge_withdraw(mm, pmd); diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 271e4432734c..75c0eef52c5d 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -40,6 +40,11 @@ int hugepages_treat_as_movable; int hugetlb_max_hstate __read_mostly; unsigned int default_hstate_idx; struct hstate hstates[HUGE_MAX_HSTATE]; +/* + * Minimum page order among possible hugepage sizes, set to a proper value + * at boot time. + */ +static unsigned int minimum_order __read_mostly = UINT_MAX; __initdata LIST_HEAD(huge_boot_pages); @@ -212,8 +217,20 @@ static inline struct hugepage_subpool *subpool_vma(struct vm_area_struct *vma) * Region tracking -- allows tracking of reservations and instantiated pages * across the pages in a mapping. * - * The region data structures are embedded into a resv_map and - * protected by a resv_map's lock + * The region data structures are embedded into a resv_map and protected + * by a resv_map's lock. The set of regions within the resv_map represent + * reservations for huge pages, or huge pages that have already been + * instantiated within the map. The from and to elements are huge page + * indicies into the associated mapping. from indicates the starting index + * of the region. to represents the first index past the end of the region. + * + * For example, a file region structure with from == 0 and to == 4 represents + * four huge pages in a mapping. It is important to note that the to element + * represents the first element past the end of the region. This is used in + * arithmetic as 4(to) - 0(from) = 4 huge pages in the region. + * + * Interval notation of the form [from, to) will be used to indicate that + * the endpoint from is inclusive and to is exclusive. */ struct file_region { struct list_head link; @@ -221,10 +238,22 @@ struct file_region { long to; }; +/* + * Add the huge page range represented by [f, t) to the reserve + * map. Existing regions will be expanded to accommodate the + * specified range. We know only existing regions need to be + * expanded, because region_add is only called after region_chg + * with the same range. If a new file_region structure must + * be allocated, it is done in region_chg. + * + * Return the number of new huge pages added to the map. This + * number is greater than or equal to zero. + */ static long region_add(struct resv_map *resv, long f, long t) { struct list_head *head = &resv->regions; struct file_region *rg, *nrg, *trg; + long add = 0; spin_lock(&resv->lock); /* Locate the region we are either in or before. */ @@ -250,16 +279,45 @@ static long region_add(struct resv_map *resv, long f, long t) if (rg->to > t) t = rg->to; if (rg != nrg) { + /* Decrement return value by the deleted range. + * Another range will span this area so that by + * end of routine add will be >= zero + */ + add -= (rg->to - rg->from); list_del(&rg->link); kfree(rg); } } + + add += (nrg->from - f); /* Added to beginning of region */ nrg->from = f; + add += t - nrg->to; /* Added to end of region */ nrg->to = t; + spin_unlock(&resv->lock); - return 0; + VM_BUG_ON(add < 0); + return add; } +/* + * Examine the existing reserve map and determine how many + * huge pages in the specified range [f, t) are NOT currently + * represented. This routine is called before a subsequent + * call to region_add that will actually modify the reserve + * map to add the specified range [f, t). region_chg does + * not change the number of huge pages represented by the + * map. However, if the existing regions in the map can not + * be expanded to represent the new range, a new file_region + * structure is added to the map as a placeholder. This is + * so that the subsequent region_add call will have all the + * regions it needs and will not fail. + * + * Returns the number of huge pages that need to be added + * to the existing reservation map for the range [f, t). + * This number is greater or equal to zero. -ENOMEM is + * returned if a new file_region structure is needed and can + * not be allocated. + */ static long region_chg(struct resv_map *resv, long f, long t) { struct list_head *head = &resv->regions; @@ -326,6 +384,11 @@ out_nrg: return chg; } +/* + * Truncate the reserve map at index 'end'. Modify/truncate any + * region which contains end. Delete any regions past end. + * Return the number of huge pages removed from the map. + */ static long region_truncate(struct resv_map *resv, long end) { struct list_head *head = &resv->regions; @@ -361,6 +424,10 @@ out: return chg; } +/* + * Count and return the number of huge pages in the reserve map + * that intersect with the range [f, t). + */ static long region_count(struct resv_map *resv, long f, long t) { struct list_head *head = &resv->regions; @@ -1188,19 +1255,13 @@ static void dissolve_free_huge_page(struct page *page) */ void dissolve_free_huge_pages(unsigned long start_pfn, unsigned long end_pfn) { - unsigned int order = 8 * sizeof(void *); unsigned long pfn; - struct hstate *h; if (!hugepages_supported()) return; - /* Set scan step to minimum hugepage size */ - for_each_hstate(h) - if (order > huge_page_order(h)) - order = huge_page_order(h); - VM_BUG_ON(!IS_ALIGNED(start_pfn, 1 << order)); - for (pfn = start_pfn; pfn < end_pfn; pfn += 1 << order) + VM_BUG_ON(!IS_ALIGNED(start_pfn, 1 << minimum_order)); + for (pfn = start_pfn; pfn < end_pfn; pfn += 1 << minimum_order) dissolve_free_huge_page(pfn_to_page(pfn)); } @@ -1423,46 +1484,56 @@ static void return_unused_surplus_pages(struct hstate *h, } /* - * Determine if the huge page at addr within the vma has an associated - * reservation. Where it does not we will need to logically increase - * reservation and actually increase subpool usage before an allocation - * can occur. Where any new reservation would be required the - * reservation change is prepared, but not committed. Once the page - * has been allocated from the subpool and instantiated the change should - * be committed via vma_commit_reservation. No action is required on - * failure. + * vma_needs_reservation and vma_commit_reservation are used by the huge + * page allocation routines to manage reservations. + * + * vma_needs_reservation is called to determine if the huge page at addr + * within the vma has an associated reservation. If a reservation is + * needed, the value 1 is returned. The caller is then responsible for + * managing the global reservation and subpool usage counts. After + * the huge page has been allocated, vma_commit_reservation is called + * to add the page to the reservation map. + * + * In the normal case, vma_commit_reservation returns the same value + * as the preceding vma_needs_reservation call. The only time this + * is not the case is if a reserve map was changed between calls. It + * is the responsibility of the caller to notice the difference and + * take appropriate action. */ -static long vma_needs_reservation(struct hstate *h, - struct vm_area_struct *vma, unsigned long addr) +static long __vma_reservation_common(struct hstate *h, + struct vm_area_struct *vma, unsigned long addr, + bool commit) { struct resv_map *resv; pgoff_t idx; - long chg; + long ret; resv = vma_resv_map(vma); if (!resv) return 1; idx = vma_hugecache_offset(h, vma, addr); - chg = region_chg(resv, idx, idx + 1); + if (commit) + ret = region_add(resv, idx, idx + 1); + else + ret = region_chg(resv, idx, idx + 1); if (vma->vm_flags & VM_MAYSHARE) - return chg; + return ret; else - return chg < 0 ? chg : 0; + return ret < 0 ? ret : 0; } -static void vma_commit_reservation(struct hstate *h, + +static long vma_needs_reservation(struct hstate *h, struct vm_area_struct *vma, unsigned long addr) { - struct resv_map *resv; - pgoff_t idx; - - resv = vma_resv_map(vma); - if (!resv) - return; + return __vma_reservation_common(h, vma, addr, false); +} - idx = vma_hugecache_offset(h, vma, addr); - region_add(resv, idx, idx + 1); +static long vma_commit_reservation(struct hstate *h, + struct vm_area_struct *vma, unsigned long addr) +{ + return __vma_reservation_common(h, vma, addr, true); } static struct page *alloc_huge_page(struct vm_area_struct *vma, @@ -1471,7 +1542,7 @@ static struct page *alloc_huge_page(struct vm_area_struct *vma, struct hugepage_subpool *spool = subpool_vma(vma); struct hstate *h = hstate_vma(vma); struct page *page; - long chg; + long chg, commit; int ret, idx; struct hugetlb_cgroup *h_cg; @@ -1512,7 +1583,22 @@ static struct page *alloc_huge_page(struct vm_area_struct *vma, set_page_private(page, (unsigned long)spool); - vma_commit_reservation(h, vma, addr); + commit = vma_commit_reservation(h, vma, addr); + if (unlikely(chg > commit)) { + /* + * The page was added to the reservation map between + * vma_needs_reservation and vma_commit_reservation. + * This indicates a race with hugetlb_reserve_pages. + * Adjust for the subpool count incremented above AND + * in hugetlb_reserve_pages for the same page. Also, + * the reservation count added in hugetlb_reserve_pages + * no longer applies. + */ + long rsv_adjust; + + rsv_adjust = hugepage_subpool_put_pages(spool, 1); + hugetlb_acct_memory(h, -rsv_adjust); + } return page; out_uncharge_cgroup: @@ -1627,10 +1713,14 @@ static void __init hugetlb_init_hstates(void) struct hstate *h; for_each_hstate(h) { + if (minimum_order > huge_page_order(h)) + minimum_order = huge_page_order(h); + /* oversize hugepages were init'ed in early boot */ if (!hstate_is_gigantic(h)) hugetlb_hstate_alloc_pages(h); } + VM_BUG_ON(minimum_order == UINT_MAX); } static char * __init memfmt(char *buf, unsigned long n) @@ -3626,8 +3716,24 @@ int hugetlb_reserve_pages(struct inode *inode, * consumed reservations are stored in the map. Hence, nothing * else has to be done for private mappings here */ - if (!vma || vma->vm_flags & VM_MAYSHARE) - region_add(resv_map, from, to); + if (!vma || vma->vm_flags & VM_MAYSHARE) { + long add = region_add(resv_map, from, to); + + if (unlikely(chg > add)) { + /* + * pages in this range were added to the reserve + * map between region_chg and region_add. This + * indicates a race with alloc_huge_page. Adjust + * the subpool and reserve counts modified above + * based on the difference. + */ + long rsv_adjust; + + rsv_adjust = hugepage_subpool_put_pages(spool, + chg - add); + hugetlb_acct_memory(h, -rsv_adjust); + } + } return 0; out_err: if (vma && is_vma_resv_set(vma, HPAGE_RESV_OWNER)) @@ -3789,6 +3895,11 @@ pte_t *huge_pmd_share(struct mm_struct *mm, unsigned long addr, pud_t *pud) { return NULL; } + +int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep) +{ + return 0; +} #define want_pmd_share() (0) #endif /* CONFIG_ARCH_WANT_HUGE_PMD_SHARE */ diff --git a/mm/hwpoison-inject.c b/mm/hwpoison-inject.c index 4ca5fe0042e1..bf73ac17dad4 100644 --- a/mm/hwpoison-inject.c +++ b/mm/hwpoison-inject.c @@ -28,7 +28,7 @@ static int hwpoison_inject(void *data, u64 val) /* * This implies unable to support free buddy pages. */ - if (!get_page_unless_zero(hpage)) + if (!get_hwpoison_page(p)) return 0; if (!hwpoison_filter_enable) @@ -58,7 +58,7 @@ inject: pr_info("Injecting memory failure at pfn %#lx\n", pfn); return memory_failure(pfn, 18, MF_COUNT_INCREASED); put_out: - put_page(hpage); + put_page(p); return 0; } diff --git a/mm/kmemleak.c b/mm/kmemleak.c index f0fe4f2c1fa7..cf79f110157c 100644 --- a/mm/kmemleak.c +++ b/mm/kmemleak.c @@ -53,6 +53,13 @@ * modifications to the memory scanning parameters including the scan_thread * pointer * + * Locks and mutexes are acquired/nested in the following order: + * + * scan_mutex [-> object->lock] -> kmemleak_lock -> other_object->lock (SINGLE_DEPTH_NESTING) + * + * No kmemleak_lock and object->lock nesting is allowed outside scan_mutex + * regions. + * * The kmemleak_object structures have a use_count incremented or decremented * using the get_object()/put_object() functions. When the use_count becomes * 0, this count can no longer be incremented and put_object() schedules the @@ -195,6 +202,8 @@ static struct kmem_cache *scan_area_cache; /* set if tracing memory operations is enabled */ static int kmemleak_enabled; +/* same as above but only for the kmemleak_free() callback */ +static int kmemleak_free_enabled; /* set in the late_initcall if there were no errors */ static int kmemleak_initialized; /* enables or disables early logging of the memory operations */ @@ -483,8 +492,7 @@ static struct kmemleak_object *find_and_get_object(unsigned long ptr, int alias) rcu_read_lock(); read_lock_irqsave(&kmemleak_lock, flags); - if (ptr >= min_addr && ptr < max_addr) - object = lookup_object(ptr, alias); + object = lookup_object(ptr, alias); read_unlock_irqrestore(&kmemleak_lock, flags); /* check whether the object is still available */ @@ -496,6 +504,27 @@ static struct kmemleak_object *find_and_get_object(unsigned long ptr, int alias) } /* + * Look up an object in the object search tree and remove it from both + * object_tree_root and object_list. The returned object's use_count should be + * at least 1, as initially set by create_object(). + */ +static struct kmemleak_object *find_and_remove_object(unsigned long ptr, int alias) +{ + unsigned long flags; + struct kmemleak_object *object; + + write_lock_irqsave(&kmemleak_lock, flags); + object = lookup_object(ptr, alias); + if (object) { + rb_erase(&object->rb_node, &object_tree_root); + list_del_rcu(&object->object_list); + } + write_unlock_irqrestore(&kmemleak_lock, flags); + + return object; +} + +/* * Save stack trace to the given array of MAX_TRACE size. */ static int __save_stack_trace(unsigned long *trace) @@ -580,11 +609,13 @@ static struct kmemleak_object *create_object(unsigned long ptr, size_t size, kmemleak_stop("Cannot insert 0x%lx into the object " "search tree (overlaps existing)\n", ptr); + /* + * No need for parent->lock here since "parent" cannot + * be freed while the kmemleak_lock is held. + */ + dump_object_info(parent); kmem_cache_free(object_cache, object); - object = parent; - spin_lock(&object->lock); - dump_object_info(object); - spin_unlock(&object->lock); + object = NULL; goto out; } } @@ -598,20 +629,14 @@ out: } /* - * Remove the metadata (struct kmemleak_object) for a memory block from the - * object_list and object_tree_root and decrement its use_count. + * Mark the object as not allocated and schedule RCU freeing via put_object(). */ static void __delete_object(struct kmemleak_object *object) { unsigned long flags; - write_lock_irqsave(&kmemleak_lock, flags); - rb_erase(&object->rb_node, &object_tree_root); - list_del_rcu(&object->object_list); - write_unlock_irqrestore(&kmemleak_lock, flags); - WARN_ON(!(object->flags & OBJECT_ALLOCATED)); - WARN_ON(atomic_read(&object->use_count) < 2); + WARN_ON(atomic_read(&object->use_count) < 1); /* * Locking here also ensures that the corresponding memory block @@ -631,7 +656,7 @@ static void delete_object_full(unsigned long ptr) { struct kmemleak_object *object; - object = find_and_get_object(ptr, 0); + object = find_and_remove_object(ptr, 0); if (!object) { #ifdef DEBUG kmemleak_warn("Freeing unknown object at 0x%08lx\n", @@ -640,7 +665,6 @@ static void delete_object_full(unsigned long ptr) return; } __delete_object(object); - put_object(object); } /* @@ -653,7 +677,7 @@ static void delete_object_part(unsigned long ptr, size_t size) struct kmemleak_object *object; unsigned long start, end; - object = find_and_get_object(ptr, 1); + object = find_and_remove_object(ptr, 1); if (!object) { #ifdef DEBUG kmemleak_warn("Partially freeing unknown object at 0x%08lx " @@ -661,7 +685,6 @@ static void delete_object_part(unsigned long ptr, size_t size) #endif return; } - __delete_object(object); /* * Create one or two objects that may result from the memory block @@ -679,7 +702,7 @@ static void delete_object_part(unsigned long ptr, size_t size) create_object(ptr + size, end - ptr - size, object->min_count, GFP_KERNEL); - put_object(object); + __delete_object(object); } static void __paint_it(struct kmemleak_object *object, int color) @@ -907,12 +930,13 @@ EXPORT_SYMBOL_GPL(kmemleak_alloc); * kmemleak_alloc_percpu - register a newly allocated __percpu object * @ptr: __percpu pointer to beginning of the object * @size: size of the object + * @gfp: flags used for kmemleak internal memory allocations * * This function is called from the kernel percpu allocator when a new object - * (memory block) is allocated (alloc_percpu). It assumes GFP_KERNEL - * allocation. + * (memory block) is allocated (alloc_percpu). */ -void __ref kmemleak_alloc_percpu(const void __percpu *ptr, size_t size) +void __ref kmemleak_alloc_percpu(const void __percpu *ptr, size_t size, + gfp_t gfp) { unsigned int cpu; @@ -925,7 +949,7 @@ void __ref kmemleak_alloc_percpu(const void __percpu *ptr, size_t size) if (kmemleak_enabled && ptr && !IS_ERR(ptr)) for_each_possible_cpu(cpu) create_object((unsigned long)per_cpu_ptr(ptr, cpu), - size, 0, GFP_KERNEL); + size, 0, gfp); else if (kmemleak_early_log) log_early(KMEMLEAK_ALLOC_PERCPU, ptr, size, 0); } @@ -942,7 +966,7 @@ void __ref kmemleak_free(const void *ptr) { pr_debug("%s(0x%p)\n", __func__, ptr); - if (kmemleak_enabled && ptr && !IS_ERR(ptr)) + if (kmemleak_free_enabled && ptr && !IS_ERR(ptr)) delete_object_full((unsigned long)ptr); else if (kmemleak_early_log) log_early(KMEMLEAK_FREE, ptr, 0, 0); @@ -982,7 +1006,7 @@ void __ref kmemleak_free_percpu(const void __percpu *ptr) pr_debug("%s(0x%p)\n", __func__, ptr); - if (kmemleak_enabled && ptr && !IS_ERR(ptr)) + if (kmemleak_free_enabled && ptr && !IS_ERR(ptr)) for_each_possible_cpu(cpu) delete_object_full((unsigned long)per_cpu_ptr(ptr, cpu)); @@ -1148,19 +1172,18 @@ static int scan_should_stop(void) * found to the gray list. */ static void scan_block(void *_start, void *_end, - struct kmemleak_object *scanned, int allow_resched) + struct kmemleak_object *scanned) { unsigned long *ptr; unsigned long *start = PTR_ALIGN(_start, BYTES_PER_POINTER); unsigned long *end = _end - (BYTES_PER_POINTER - 1); + unsigned long flags; + read_lock_irqsave(&kmemleak_lock, flags); for (ptr = start; ptr < end; ptr++) { struct kmemleak_object *object; - unsigned long flags; unsigned long pointer; - if (allow_resched) - cond_resched(); if (scan_should_stop()) break; @@ -1173,26 +1196,31 @@ static void scan_block(void *_start, void *_end, pointer = *ptr; kasan_enable_current(); - object = find_and_get_object(pointer, 1); + if (pointer < min_addr || pointer >= max_addr) + continue; + + /* + * No need for get_object() here since we hold kmemleak_lock. + * object->use_count cannot be dropped to 0 while the object + * is still present in object_tree_root and object_list + * (with updates protected by kmemleak_lock). + */ + object = lookup_object(pointer, 1); if (!object) continue; - if (object == scanned) { + if (object == scanned) /* self referenced, ignore */ - put_object(object); continue; - } /* * Avoid the lockdep recursive warning on object->lock being * previously acquired in scan_object(). These locks are * enclosed by scan_mutex. */ - spin_lock_irqsave_nested(&object->lock, flags, - SINGLE_DEPTH_NESTING); + spin_lock_nested(&object->lock, SINGLE_DEPTH_NESTING); if (!color_white(object)) { /* non-orphan, ignored or new */ - spin_unlock_irqrestore(&object->lock, flags); - put_object(object); + spin_unlock(&object->lock); continue; } @@ -1204,13 +1232,27 @@ static void scan_block(void *_start, void *_end, */ object->count++; if (color_gray(object)) { + /* put_object() called when removing from gray_list */ + WARN_ON(!get_object(object)); list_add_tail(&object->gray_list, &gray_list); - spin_unlock_irqrestore(&object->lock, flags); - continue; } + spin_unlock(&object->lock); + } + read_unlock_irqrestore(&kmemleak_lock, flags); +} - spin_unlock_irqrestore(&object->lock, flags); - put_object(object); +/* + * Scan a large memory block in MAX_SCAN_SIZE chunks to reduce the latency. + */ +static void scan_large_block(void *start, void *end) +{ + void *next; + + while (start < end) { + next = min(start + MAX_SCAN_SIZE, end); + scan_block(start, next, NULL); + start = next; + cond_resched(); } } @@ -1236,22 +1278,25 @@ static void scan_object(struct kmemleak_object *object) if (hlist_empty(&object->area_list)) { void *start = (void *)object->pointer; void *end = (void *)(object->pointer + object->size); + void *next; - while (start < end && (object->flags & OBJECT_ALLOCATED) && - !(object->flags & OBJECT_NO_SCAN)) { - scan_block(start, min(start + MAX_SCAN_SIZE, end), - object, 0); - start += MAX_SCAN_SIZE; + do { + next = min(start + MAX_SCAN_SIZE, end); + scan_block(start, next, object); + + start = next; + if (start >= end) + break; spin_unlock_irqrestore(&object->lock, flags); cond_resched(); spin_lock_irqsave(&object->lock, flags); - } + } while (object->flags & OBJECT_ALLOCATED); } else hlist_for_each_entry(area, &object->area_list, node) scan_block((void *)area->start, (void *)(area->start + area->size), - object, 0); + object); out: spin_unlock_irqrestore(&object->lock, flags); } @@ -1328,14 +1373,14 @@ static void kmemleak_scan(void) rcu_read_unlock(); /* data/bss scanning */ - scan_block(_sdata, _edata, NULL, 1); - scan_block(__bss_start, __bss_stop, NULL, 1); + scan_large_block(_sdata, _edata); + scan_large_block(__bss_start, __bss_stop); #ifdef CONFIG_SMP /* per-cpu sections scanning */ for_each_possible_cpu(i) - scan_block(__per_cpu_start + per_cpu_offset(i), - __per_cpu_end + per_cpu_offset(i), NULL, 1); + scan_large_block(__per_cpu_start + per_cpu_offset(i), + __per_cpu_end + per_cpu_offset(i)); #endif /* @@ -1356,7 +1401,7 @@ static void kmemleak_scan(void) /* only scan if page is in use */ if (page_count(page) == 0) continue; - scan_block(page, page + 1, NULL, 1); + scan_block(page, page + 1, NULL); } } put_online_mems(); @@ -1370,7 +1415,7 @@ static void kmemleak_scan(void) read_lock(&tasklist_lock); do_each_thread(g, p) { scan_block(task_stack_page(p), task_stack_page(p) + - THREAD_SIZE, NULL, 0); + THREAD_SIZE, NULL); } while_each_thread(g, p); read_unlock(&tasklist_lock); } @@ -1747,15 +1792,20 @@ static void __kmemleak_do_cleanup(void) */ static void kmemleak_do_cleanup(struct work_struct *work) { - mutex_lock(&scan_mutex); stop_scan_thread(); + /* + * Once the scan thread has stopped, it is safe to no longer track + * object freeing. Ordering of the scan thread stopping and the memory + * accesses below is guaranteed by the kthread_stop() function. + */ + kmemleak_free_enabled = 0; + if (!kmemleak_found_leaks) __kmemleak_do_cleanup(); else pr_info("Kmemleak disabled without freeing internal data. " "Reclaim the memory with \"echo clear > /sys/kernel/debug/kmemleak\"\n"); - mutex_unlock(&scan_mutex); } static DECLARE_WORK(cleanup_work, kmemleak_do_cleanup); @@ -1776,6 +1826,8 @@ static void kmemleak_disable(void) /* check whether it is too early for a kernel thread */ if (kmemleak_initialized) schedule_work(&cleanup_work); + else + kmemleak_free_enabled = 0; pr_info("Kernel memory leak detector disabled\n"); } @@ -1840,8 +1892,10 @@ void __init kmemleak_init(void) if (kmemleak_error) { local_irq_restore(flags); return; - } else + } else { kmemleak_enabled = 1; + kmemleak_free_enabled = 1; + } local_irq_restore(flags); /* diff --git a/mm/memblock.c b/mm/memblock.c index 9318b567ed79..1b444c730846 100644 --- a/mm/memblock.c +++ b/mm/memblock.c @@ -54,10 +54,16 @@ int memblock_debug __initdata_memblock; #ifdef CONFIG_MOVABLE_NODE bool movable_node_enabled __initdata_memblock = false; #endif +static bool system_has_some_mirror __initdata_memblock = false; static int memblock_can_resize __initdata_memblock; static int memblock_memory_in_slab __initdata_memblock = 0; static int memblock_reserved_in_slab __initdata_memblock = 0; +ulong __init_memblock choose_memblock_flags(void) +{ + return system_has_some_mirror ? MEMBLOCK_MIRROR : MEMBLOCK_NONE; +} + /* inline so we don't get a warning when pr_debug is compiled out */ static __init_memblock const char * memblock_type_name(struct memblock_type *type) @@ -107,6 +113,7 @@ static long __init_memblock memblock_overlaps_region(struct memblock_type *type, * @size: size of free area to find * @align: alignment of free area to find * @nid: nid of the free area to find, %NUMA_NO_NODE for any node + * @flags: pick from blocks based on memory attributes * * Utility called from memblock_find_in_range_node(), find free area bottom-up. * @@ -115,12 +122,13 @@ static long __init_memblock memblock_overlaps_region(struct memblock_type *type, */ static phys_addr_t __init_memblock __memblock_find_range_bottom_up(phys_addr_t start, phys_addr_t end, - phys_addr_t size, phys_addr_t align, int nid) + phys_addr_t size, phys_addr_t align, int nid, + ulong flags) { phys_addr_t this_start, this_end, cand; u64 i; - for_each_free_mem_range(i, nid, &this_start, &this_end, NULL) { + for_each_free_mem_range(i, nid, flags, &this_start, &this_end, NULL) { this_start = clamp(this_start, start, end); this_end = clamp(this_end, start, end); @@ -139,6 +147,7 @@ __memblock_find_range_bottom_up(phys_addr_t start, phys_addr_t end, * @size: size of free area to find * @align: alignment of free area to find * @nid: nid of the free area to find, %NUMA_NO_NODE for any node + * @flags: pick from blocks based on memory attributes * * Utility called from memblock_find_in_range_node(), find free area top-down. * @@ -147,12 +156,14 @@ __memblock_find_range_bottom_up(phys_addr_t start, phys_addr_t end, */ static phys_addr_t __init_memblock __memblock_find_range_top_down(phys_addr_t start, phys_addr_t end, - phys_addr_t size, phys_addr_t align, int nid) + phys_addr_t size, phys_addr_t align, int nid, + ulong flags) { phys_addr_t this_start, this_end, cand; u64 i; - for_each_free_mem_range_reverse(i, nid, &this_start, &this_end, NULL) { + for_each_free_mem_range_reverse(i, nid, flags, &this_start, &this_end, + NULL) { this_start = clamp(this_start, start, end); this_end = clamp(this_end, start, end); @@ -174,6 +185,7 @@ __memblock_find_range_top_down(phys_addr_t start, phys_addr_t end, * @start: start of candidate range * @end: end of candidate range, can be %MEMBLOCK_ALLOC_{ANYWHERE|ACCESSIBLE} * @nid: nid of the free area to find, %NUMA_NO_NODE for any node + * @flags: pick from blocks based on memory attributes * * Find @size free area aligned to @align in the specified range and node. * @@ -190,7 +202,7 @@ __memblock_find_range_top_down(phys_addr_t start, phys_addr_t end, */ phys_addr_t __init_memblock memblock_find_in_range_node(phys_addr_t size, phys_addr_t align, phys_addr_t start, - phys_addr_t end, int nid) + phys_addr_t end, int nid, ulong flags) { phys_addr_t kernel_end, ret; @@ -215,7 +227,7 @@ phys_addr_t __init_memblock memblock_find_in_range_node(phys_addr_t size, /* ok, try bottom-up allocation first */ ret = __memblock_find_range_bottom_up(bottom_up_start, end, - size, align, nid); + size, align, nid, flags); if (ret) return ret; @@ -233,7 +245,8 @@ phys_addr_t __init_memblock memblock_find_in_range_node(phys_addr_t size, "memory hotunplug may be affected\n"); } - return __memblock_find_range_top_down(start, end, size, align, nid); + return __memblock_find_range_top_down(start, end, size, align, nid, + flags); } /** @@ -252,8 +265,21 @@ phys_addr_t __init_memblock memblock_find_in_range(phys_addr_t start, phys_addr_t end, phys_addr_t size, phys_addr_t align) { - return memblock_find_in_range_node(size, align, start, end, - NUMA_NO_NODE); + phys_addr_t ret; + ulong flags = choose_memblock_flags(); + +again: + ret = memblock_find_in_range_node(size, align, start, end, + NUMA_NO_NODE, flags); + + if (!ret && (flags & MEMBLOCK_MIRROR)) { + pr_warn("Could not allocate %pap bytes of mirrored memory\n", + &size); + flags &= ~MEMBLOCK_MIRROR; + goto again; + } + + return ret; } static void __init_memblock memblock_remove_region(struct memblock_type *type, unsigned long r) @@ -779,9 +805,25 @@ int __init_memblock memblock_clear_hotplug(phys_addr_t base, phys_addr_t size) } /** + * memblock_mark_mirror - Mark mirrored memory with flag MEMBLOCK_MIRROR. + * @base: the base phys addr of the region + * @size: the size of the region + * + * Return 0 on succees, -errno on failure. + */ +int __init_memblock memblock_mark_mirror(phys_addr_t base, phys_addr_t size) +{ + system_has_some_mirror = true; + + return memblock_setclr_flag(base, size, 1, MEMBLOCK_MIRROR); +} + + +/** * __next__mem_range - next function for for_each_free_mem_range() etc. * @idx: pointer to u64 loop variable * @nid: node selector, %NUMA_NO_NODE for all nodes + * @flags: pick from blocks based on memory attributes * @type_a: pointer to memblock_type from where the range is taken * @type_b: pointer to memblock_type which excludes memory from being taken * @out_start: ptr to phys_addr_t for start address of the range, can be %NULL @@ -803,7 +845,7 @@ int __init_memblock memblock_clear_hotplug(phys_addr_t base, phys_addr_t size) * As both region arrays are sorted, the function advances the two indices * in lockstep and returns each intersection. */ -void __init_memblock __next_mem_range(u64 *idx, int nid, +void __init_memblock __next_mem_range(u64 *idx, int nid, ulong flags, struct memblock_type *type_a, struct memblock_type *type_b, phys_addr_t *out_start, @@ -831,6 +873,10 @@ void __init_memblock __next_mem_range(u64 *idx, int nid, if (movable_node_is_enabled() && memblock_is_hotpluggable(m)) continue; + /* if we want mirror memory skip non-mirror memory regions */ + if ((flags & MEMBLOCK_MIRROR) && !memblock_is_mirror(m)) + continue; + if (!type_b) { if (out_start) *out_start = m_start; @@ -895,6 +941,7 @@ void __init_memblock __next_mem_range(u64 *idx, int nid, * * @idx: pointer to u64 loop variable * @nid: nid: node selector, %NUMA_NO_NODE for all nodes + * @flags: pick from blocks based on memory attributes * @type_a: pointer to memblock_type from where the range is taken * @type_b: pointer to memblock_type which excludes memory from being taken * @out_start: ptr to phys_addr_t for start address of the range, can be %NULL @@ -903,7 +950,7 @@ void __init_memblock __next_mem_range(u64 *idx, int nid, * * Reverse of __next_mem_range(). */ -void __init_memblock __next_mem_range_rev(u64 *idx, int nid, +void __init_memblock __next_mem_range_rev(u64 *idx, int nid, ulong flags, struct memblock_type *type_a, struct memblock_type *type_b, phys_addr_t *out_start, @@ -935,6 +982,10 @@ void __init_memblock __next_mem_range_rev(u64 *idx, int nid, if (movable_node_is_enabled() && memblock_is_hotpluggable(m)) continue; + /* if we want mirror memory skip non-mirror memory regions */ + if ((flags & MEMBLOCK_MIRROR) && !memblock_is_mirror(m)) + continue; + if (!type_b) { if (out_start) *out_start = m_start; @@ -1050,14 +1101,15 @@ int __init_memblock memblock_set_node(phys_addr_t base, phys_addr_t size, static phys_addr_t __init memblock_alloc_range_nid(phys_addr_t size, phys_addr_t align, phys_addr_t start, - phys_addr_t end, int nid) + phys_addr_t end, int nid, ulong flags) { phys_addr_t found; if (!align) align = SMP_CACHE_BYTES; - found = memblock_find_in_range_node(size, align, start, end, nid); + found = memblock_find_in_range_node(size, align, start, end, nid, + flags); if (found && !memblock_reserve(found, size)) { /* * The min_count is set to 0 so that memblock allocations are @@ -1070,26 +1122,40 @@ static phys_addr_t __init memblock_alloc_range_nid(phys_addr_t size, } phys_addr_t __init memblock_alloc_range(phys_addr_t size, phys_addr_t align, - phys_addr_t start, phys_addr_t end) + phys_addr_t start, phys_addr_t end, + ulong flags) { - return memblock_alloc_range_nid(size, align, start, end, NUMA_NO_NODE); + return memblock_alloc_range_nid(size, align, start, end, NUMA_NO_NODE, + flags); } static phys_addr_t __init memblock_alloc_base_nid(phys_addr_t size, phys_addr_t align, phys_addr_t max_addr, - int nid) + int nid, ulong flags) { - return memblock_alloc_range_nid(size, align, 0, max_addr, nid); + return memblock_alloc_range_nid(size, align, 0, max_addr, nid, flags); } phys_addr_t __init memblock_alloc_nid(phys_addr_t size, phys_addr_t align, int nid) { - return memblock_alloc_base_nid(size, align, MEMBLOCK_ALLOC_ACCESSIBLE, nid); + ulong flags = choose_memblock_flags(); + phys_addr_t ret; + +again: + ret = memblock_alloc_base_nid(size, align, MEMBLOCK_ALLOC_ACCESSIBLE, + nid, flags); + + if (!ret && (flags & MEMBLOCK_MIRROR)) { + flags &= ~MEMBLOCK_MIRROR; + goto again; + } + return ret; } phys_addr_t __init __memblock_alloc_base(phys_addr_t size, phys_addr_t align, phys_addr_t max_addr) { - return memblock_alloc_base_nid(size, align, max_addr, NUMA_NO_NODE); + return memblock_alloc_base_nid(size, align, max_addr, NUMA_NO_NODE, + MEMBLOCK_NONE); } phys_addr_t __init memblock_alloc_base(phys_addr_t size, phys_addr_t align, phys_addr_t max_addr) @@ -1153,6 +1219,7 @@ static void * __init memblock_virt_alloc_internal( { phys_addr_t alloc; void *ptr; + ulong flags = choose_memblock_flags(); if (WARN_ONCE(nid == MAX_NUMNODES, "Usage of MAX_NUMNODES is deprecated. Use NUMA_NO_NODE instead\n")) nid = NUMA_NO_NODE; @@ -1173,13 +1240,14 @@ static void * __init memblock_virt_alloc_internal( again: alloc = memblock_find_in_range_node(size, align, min_addr, max_addr, - nid); + nid, flags); if (alloc) goto done; if (nid != NUMA_NO_NODE) { alloc = memblock_find_in_range_node(size, align, min_addr, - max_addr, NUMA_NO_NODE); + max_addr, NUMA_NO_NODE, + flags); if (alloc) goto done; } @@ -1187,10 +1255,16 @@ again: if (min_addr) { min_addr = 0; goto again; - } else { - goto error; } + if (flags & MEMBLOCK_MIRROR) { + flags &= ~MEMBLOCK_MIRROR; + pr_warn("Could not allocate %pap bytes of mirrored memory\n", + &size); + goto again; + } + + return NULL; done: memblock_reserve(alloc, size); ptr = phys_to_virt(alloc); @@ -1205,9 +1279,6 @@ done: kmemleak_alloc(ptr, size, 0, 0); return ptr; - -error: - return NULL; } /** diff --git a/mm/memcontrol.c b/mm/memcontrol.c index a04225d372ba..e65f7b0131d3 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -285,9 +285,9 @@ struct mem_cgroup { */ bool use_hierarchy; + /* protected by memcg_oom_lock */ bool oom_lock; - atomic_t under_oom; - atomic_t oom_wakeups; + int under_oom; int swappiness; /* OOM-Killer disable */ @@ -1530,14 +1530,16 @@ static void mem_cgroup_out_of_memory(struct mem_cgroup *memcg, gfp_t gfp_mask, unsigned int points = 0; struct task_struct *chosen = NULL; + mutex_lock(&oom_lock); + /* * If current has a pending SIGKILL or is exiting, then automatically * select it. The goal is to allow it to allocate so that it may * quickly exit and free its memory. */ if (fatal_signal_pending(current) || task_will_free_mem(current)) { - mark_tsk_oom_victim(current); - return; + mark_oom_victim(current); + goto unlock; } check_panic_on_oom(CONSTRAINT_MEMCG, gfp_mask, order, NULL, memcg); @@ -1564,7 +1566,7 @@ static void mem_cgroup_out_of_memory(struct mem_cgroup *memcg, gfp_t gfp_mask, mem_cgroup_iter_break(memcg, iter); if (chosen) put_task_struct(chosen); - return; + goto unlock; case OOM_SCAN_OK: break; }; @@ -1585,11 +1587,13 @@ static void mem_cgroup_out_of_memory(struct mem_cgroup *memcg, gfp_t gfp_mask, css_task_iter_end(&it); } - if (!chosen) - return; - points = chosen_points * 1000 / totalpages; - oom_kill_process(chosen, gfp_mask, order, points, totalpages, memcg, - NULL, "Memory cgroup out of memory"); + if (chosen) { + points = chosen_points * 1000 / totalpages; + oom_kill_process(chosen, gfp_mask, order, points, totalpages, + memcg, NULL, "Memory cgroup out of memory"); + } +unlock: + mutex_unlock(&oom_lock); } #if MAX_NUMNODES > 1 @@ -1806,8 +1810,10 @@ static void mem_cgroup_mark_under_oom(struct mem_cgroup *memcg) { struct mem_cgroup *iter; + spin_lock(&memcg_oom_lock); for_each_mem_cgroup_tree(iter, memcg) - atomic_inc(&iter->under_oom); + iter->under_oom++; + spin_unlock(&memcg_oom_lock); } static void mem_cgroup_unmark_under_oom(struct mem_cgroup *memcg) @@ -1816,11 +1822,13 @@ static void mem_cgroup_unmark_under_oom(struct mem_cgroup *memcg) /* * When a new child is created while the hierarchy is under oom, - * mem_cgroup_oom_lock() may not be called. We have to use - * atomic_add_unless() here. + * mem_cgroup_oom_lock() may not be called. Watch for underflow. */ + spin_lock(&memcg_oom_lock); for_each_mem_cgroup_tree(iter, memcg) - atomic_add_unless(&iter->under_oom, -1, 0); + if (iter->under_oom > 0) + iter->under_oom--; + spin_unlock(&memcg_oom_lock); } static DECLARE_WAIT_QUEUE_HEAD(memcg_oom_waitq); @@ -1846,17 +1854,18 @@ static int memcg_oom_wake_function(wait_queue_t *wait, return autoremove_wake_function(wait, mode, sync, arg); } -static void memcg_wakeup_oom(struct mem_cgroup *memcg) -{ - atomic_inc(&memcg->oom_wakeups); - /* for filtering, pass "memcg" as argument. */ - __wake_up(&memcg_oom_waitq, TASK_NORMAL, 0, memcg); -} - static void memcg_oom_recover(struct mem_cgroup *memcg) { - if (memcg && atomic_read(&memcg->under_oom)) - memcg_wakeup_oom(memcg); + /* + * For the following lockless ->under_oom test, the only required + * guarantee is that it must see the state asserted by an OOM when + * this function is called as a result of userland actions + * triggered by the notification of the OOM. This is trivially + * achieved by invoking mem_cgroup_mark_under_oom() before + * triggering notification. + */ + if (memcg && memcg->under_oom) + __wake_up(&memcg_oom_waitq, TASK_NORMAL, 0, memcg); } static void mem_cgroup_oom(struct mem_cgroup *memcg, gfp_t mask, int order) @@ -3864,7 +3873,7 @@ static int mem_cgroup_oom_register_event(struct mem_cgroup *memcg, list_add(&event->list, &memcg->oom_notify); /* already in OOM ? */ - if (atomic_read(&memcg->under_oom)) + if (memcg->under_oom) eventfd_signal(eventfd, 1); spin_unlock(&memcg_oom_lock); @@ -3893,7 +3902,7 @@ static int mem_cgroup_oom_control_read(struct seq_file *sf, void *v) struct mem_cgroup *memcg = mem_cgroup_from_css(seq_css(sf)); seq_printf(sf, "oom_kill_disable %d\n", memcg->oom_kill_disable); - seq_printf(sf, "under_oom %d\n", (bool)atomic_read(&memcg->under_oom)); + seq_printf(sf, "under_oom %d\n", (bool)memcg->under_oom); return 0; } diff --git a/mm/memory-failure.c b/mm/memory-failure.c index 501820c815b3..c53543d89282 100644 --- a/mm/memory-failure.c +++ b/mm/memory-failure.c @@ -20,6 +20,14 @@ * this code has to be extremely careful. Generally it tries to use * normal locking rules, as in get the standard locks, even if that means * the error handling takes potentially a long time. + * + * It can be very tempting to add handling for obscure cases here. + * In general any code for handling new cases should only be added iff: + * - You know how to test it. + * - You have a test that can be added to mce-test + * https://git.kernel.org/cgit/utils/cpu/mce/mce-test.git/ + * - The case actually shows up as a frequent (top 10) page state in + * tools/vm/page-types when running a real workload. * * There are several operations here with exponential complexity because * of unsuitable VM data structures. For example the operation to map back @@ -28,13 +36,6 @@ * are rare we hope to get away with this. This avoids impacting the core * VM. */ - -/* - * Notebook: - * - hugetlb needs more code - * - kcore/oldmem/vmcore/mem/kmem check for hwpoison pages - * - pass bad pages to kdump next kernel - */ #include <linux/kernel.h> #include <linux/mm.h> #include <linux/page-flags.h> @@ -56,6 +57,7 @@ #include <linux/mm_inline.h> #include <linux/kfifo.h> #include "internal.h" +#include "ras/ras_event.h" int sysctl_memory_failure_early_kill __read_mostly = 0; @@ -503,68 +505,34 @@ static void collect_procs(struct page *page, struct list_head *tokill, kfree(tk); } -/* - * Error handlers for various types of pages. - */ - -enum outcome { - IGNORED, /* Error: cannot be handled */ - FAILED, /* Error: handling failed */ - DELAYED, /* Will be handled later */ - RECOVERED, /* Successfully recovered */ -}; - static const char *action_name[] = { - [IGNORED] = "Ignored", - [FAILED] = "Failed", - [DELAYED] = "Delayed", - [RECOVERED] = "Recovered", -}; - -enum action_page_type { - MSG_KERNEL, - MSG_KERNEL_HIGH_ORDER, - MSG_SLAB, - MSG_DIFFERENT_COMPOUND, - MSG_POISONED_HUGE, - MSG_HUGE, - MSG_FREE_HUGE, - MSG_UNMAP_FAILED, - MSG_DIRTY_SWAPCACHE, - MSG_CLEAN_SWAPCACHE, - MSG_DIRTY_MLOCKED_LRU, - MSG_CLEAN_MLOCKED_LRU, - MSG_DIRTY_UNEVICTABLE_LRU, - MSG_CLEAN_UNEVICTABLE_LRU, - MSG_DIRTY_LRU, - MSG_CLEAN_LRU, - MSG_TRUNCATED_LRU, - MSG_BUDDY, - MSG_BUDDY_2ND, - MSG_UNKNOWN, + [MF_IGNORED] = "Ignored", + [MF_FAILED] = "Failed", + [MF_DELAYED] = "Delayed", + [MF_RECOVERED] = "Recovered", }; static const char * const action_page_types[] = { - [MSG_KERNEL] = "reserved kernel page", - [MSG_KERNEL_HIGH_ORDER] = "high-order kernel page", - [MSG_SLAB] = "kernel slab page", - [MSG_DIFFERENT_COMPOUND] = "different compound page after locking", - [MSG_POISONED_HUGE] = "huge page already hardware poisoned", - [MSG_HUGE] = "huge page", - [MSG_FREE_HUGE] = "free huge page", - [MSG_UNMAP_FAILED] = "unmapping failed page", - [MSG_DIRTY_SWAPCACHE] = "dirty swapcache page", - [MSG_CLEAN_SWAPCACHE] = "clean swapcache page", - [MSG_DIRTY_MLOCKED_LRU] = "dirty mlocked LRU page", - [MSG_CLEAN_MLOCKED_LRU] = "clean mlocked LRU page", - [MSG_DIRTY_UNEVICTABLE_LRU] = "dirty unevictable LRU page", - [MSG_CLEAN_UNEVICTABLE_LRU] = "clean unevictable LRU page", - [MSG_DIRTY_LRU] = "dirty LRU page", - [MSG_CLEAN_LRU] = "clean LRU page", - [MSG_TRUNCATED_LRU] = "already truncated LRU page", - [MSG_BUDDY] = "free buddy page", - [MSG_BUDDY_2ND] = "free buddy page (2nd try)", - [MSG_UNKNOWN] = "unknown page", + [MF_MSG_KERNEL] = "reserved kernel page", + [MF_MSG_KERNEL_HIGH_ORDER] = "high-order kernel page", + [MF_MSG_SLAB] = "kernel slab page", + [MF_MSG_DIFFERENT_COMPOUND] = "different compound page after locking", + [MF_MSG_POISONED_HUGE] = "huge page already hardware poisoned", + [MF_MSG_HUGE] = "huge page", + [MF_MSG_FREE_HUGE] = "free huge page", + [MF_MSG_UNMAP_FAILED] = "unmapping failed page", + [MF_MSG_DIRTY_SWAPCACHE] = "dirty swapcache page", + [MF_MSG_CLEAN_SWAPCACHE] = "clean swapcache page", + [MF_MSG_DIRTY_MLOCKED_LRU] = "dirty mlocked LRU page", + [MF_MSG_CLEAN_MLOCKED_LRU] = "clean mlocked LRU page", + [MF_MSG_DIRTY_UNEVICTABLE_LRU] = "dirty unevictable LRU page", + [MF_MSG_CLEAN_UNEVICTABLE_LRU] = "clean unevictable LRU page", + [MF_MSG_DIRTY_LRU] = "dirty LRU page", + [MF_MSG_CLEAN_LRU] = "clean LRU page", + [MF_MSG_TRUNCATED_LRU] = "already truncated LRU page", + [MF_MSG_BUDDY] = "free buddy page", + [MF_MSG_BUDDY_2ND] = "free buddy page (2nd try)", + [MF_MSG_UNKNOWN] = "unknown page", }; /* @@ -598,7 +566,7 @@ static int delete_from_lru_cache(struct page *p) */ static int me_kernel(struct page *p, unsigned long pfn) { - return IGNORED; + return MF_IGNORED; } /* @@ -607,7 +575,7 @@ static int me_kernel(struct page *p, unsigned long pfn) static int me_unknown(struct page *p, unsigned long pfn) { printk(KERN_ERR "MCE %#lx: Unknown page state\n", pfn); - return FAILED; + return MF_FAILED; } /* @@ -616,7 +584,7 @@ static int me_unknown(struct page *p, unsigned long pfn) static int me_pagecache_clean(struct page *p, unsigned long pfn) { int err; - int ret = FAILED; + int ret = MF_FAILED; struct address_space *mapping; delete_from_lru_cache(p); @@ -626,7 +594,7 @@ static int me_pagecache_clean(struct page *p, unsigned long pfn) * should be the one m_f() holds. */ if (PageAnon(p)) - return RECOVERED; + return MF_RECOVERED; /* * Now truncate the page in the page cache. This is really @@ -640,7 +608,7 @@ static int me_pagecache_clean(struct page *p, unsigned long pfn) /* * Page has been teared down in the meanwhile */ - return FAILED; + return MF_FAILED; } /* @@ -657,7 +625,7 @@ static int me_pagecache_clean(struct page *p, unsigned long pfn) !try_to_release_page(p, GFP_NOIO)) { pr_info("MCE %#lx: failed to release buffers\n", pfn); } else { - ret = RECOVERED; + ret = MF_RECOVERED; } } else { /* @@ -665,7 +633,7 @@ static int me_pagecache_clean(struct page *p, unsigned long pfn) * This fails on dirty or anything with private pages */ if (invalidate_inode_page(p)) - ret = RECOVERED; + ret = MF_RECOVERED; else printk(KERN_INFO "MCE %#lx: Failed to invalidate\n", pfn); @@ -751,9 +719,9 @@ static int me_swapcache_dirty(struct page *p, unsigned long pfn) ClearPageUptodate(p); if (!delete_from_lru_cache(p)) - return DELAYED; + return MF_DELAYED; else - return FAILED; + return MF_FAILED; } static int me_swapcache_clean(struct page *p, unsigned long pfn) @@ -761,9 +729,9 @@ static int me_swapcache_clean(struct page *p, unsigned long pfn) delete_from_swap_cache(p); if (!delete_from_lru_cache(p)) - return RECOVERED; + return MF_RECOVERED; else - return FAILED; + return MF_FAILED; } /* @@ -776,6 +744,10 @@ static int me_huge_page(struct page *p, unsigned long pfn) { int res = 0; struct page *hpage = compound_head(p); + + if (!PageHuge(hpage)) + return MF_DELAYED; + /* * We can safely recover from error on free or reserved (i.e. * not in-use) hugepage by dequeuing it from freelist. @@ -789,9 +761,9 @@ static int me_huge_page(struct page *p, unsigned long pfn) if (!(page_mapping(hpage) || PageAnon(hpage))) { res = dequeue_hwpoisoned_huge_page(hpage); if (!res) - return RECOVERED; + return MF_RECOVERED; } - return DELAYED; + return MF_DELAYED; } /* @@ -823,10 +795,10 @@ static int me_huge_page(struct page *p, unsigned long pfn) static struct page_state { unsigned long mask; unsigned long res; - enum action_page_type type; + enum mf_action_page_type type; int (*action)(struct page *p, unsigned long pfn); } error_states[] = { - { reserved, reserved, MSG_KERNEL, me_kernel }, + { reserved, reserved, MF_MSG_KERNEL, me_kernel }, /* * free pages are specially detected outside this table: * PG_buddy pages only make a small fraction of all free pages. @@ -837,31 +809,31 @@ static struct page_state { * currently unused objects without touching them. But just * treat it as standard kernel for now. */ - { slab, slab, MSG_SLAB, me_kernel }, + { slab, slab, MF_MSG_SLAB, me_kernel }, #ifdef CONFIG_PAGEFLAGS_EXTENDED - { head, head, MSG_HUGE, me_huge_page }, - { tail, tail, MSG_HUGE, me_huge_page }, + { head, head, MF_MSG_HUGE, me_huge_page }, + { tail, tail, MF_MSG_HUGE, me_huge_page }, #else - { compound, compound, MSG_HUGE, me_huge_page }, + { compound, compound, MF_MSG_HUGE, me_huge_page }, #endif - { sc|dirty, sc|dirty, MSG_DIRTY_SWAPCACHE, me_swapcache_dirty }, - { sc|dirty, sc, MSG_CLEAN_SWAPCACHE, me_swapcache_clean }, + { sc|dirty, sc|dirty, MF_MSG_DIRTY_SWAPCACHE, me_swapcache_dirty }, + { sc|dirty, sc, MF_MSG_CLEAN_SWAPCACHE, me_swapcache_clean }, - { mlock|dirty, mlock|dirty, MSG_DIRTY_MLOCKED_LRU, me_pagecache_dirty }, - { mlock|dirty, mlock, MSG_CLEAN_MLOCKED_LRU, me_pagecache_clean }, + { mlock|dirty, mlock|dirty, MF_MSG_DIRTY_MLOCKED_LRU, me_pagecache_dirty }, + { mlock|dirty, mlock, MF_MSG_CLEAN_MLOCKED_LRU, me_pagecache_clean }, - { unevict|dirty, unevict|dirty, MSG_DIRTY_UNEVICTABLE_LRU, me_pagecache_dirty }, - { unevict|dirty, unevict, MSG_CLEAN_UNEVICTABLE_LRU, me_pagecache_clean }, + { unevict|dirty, unevict|dirty, MF_MSG_DIRTY_UNEVICTABLE_LRU, me_pagecache_dirty }, + { unevict|dirty, unevict, MF_MSG_CLEAN_UNEVICTABLE_LRU, me_pagecache_clean }, - { lru|dirty, lru|dirty, MSG_DIRTY_LRU, me_pagecache_dirty }, - { lru|dirty, lru, MSG_CLEAN_LRU, me_pagecache_clean }, + { lru|dirty, lru|dirty, MF_MSG_DIRTY_LRU, me_pagecache_dirty }, + { lru|dirty, lru, MF_MSG_CLEAN_LRU, me_pagecache_clean }, /* * Catchall entry: must be at end. */ - { 0, 0, MSG_UNKNOWN, me_unknown }, + { 0, 0, MF_MSG_UNKNOWN, me_unknown }, }; #undef dirty @@ -881,8 +853,11 @@ static struct page_state { * "Dirty/Clean" indication is not 100% accurate due to the possibility of * setting PG_dirty outside page lock. See also comment above set_page_dirty(). */ -static void action_result(unsigned long pfn, enum action_page_type type, int result) +static void action_result(unsigned long pfn, enum mf_action_page_type type, + enum mf_result result) { + trace_memory_failure_event(pfn, type, result); + pr_err("MCE %#lx: recovery action for %s: %s\n", pfn, action_page_types[type], action_name[result]); } @@ -896,13 +871,13 @@ static int page_action(struct page_state *ps, struct page *p, result = ps->action(p, pfn); count = page_count(p) - 1; - if (ps->action == me_swapcache_dirty && result == DELAYED) + if (ps->action == me_swapcache_dirty && result == MF_DELAYED) count--; if (count != 0) { printk(KERN_ERR "MCE %#lx: %s still referenced by %d users\n", pfn, action_page_types[ps->type], count); - result = FAILED; + result = MF_FAILED; } action_result(pfn, ps->type, result); @@ -911,9 +886,42 @@ static int page_action(struct page_state *ps, struct page *p, * Could adjust zone counters here to correct for the missing page. */ - return (result == RECOVERED || result == DELAYED) ? 0 : -EBUSY; + return (result == MF_RECOVERED || result == MF_DELAYED) ? 0 : -EBUSY; } +/** + * get_hwpoison_page() - Get refcount for memory error handling: + * @page: raw error page (hit by memory error) + * + * Return: return 0 if failed to grab the refcount, otherwise true (some + * non-zero value.) + */ +int get_hwpoison_page(struct page *page) +{ + struct page *head = compound_head(page); + + if (PageHuge(head)) + return get_page_unless_zero(head); + + /* + * Thp tail page has special refcounting rule (refcount of tail pages + * is stored in ->_mapcount,) so we can't call get_page_unless_zero() + * directly for tail pages. + */ + if (PageTransHuge(head)) { + if (get_page_unless_zero(head)) { + if (PageTail(page)) + get_page(page); + return 1; + } else { + return 0; + } + } + + return get_page_unless_zero(page); +} +EXPORT_SYMBOL_GPL(get_hwpoison_page); + /* * Do all that is necessary to remove user space mappings. Unmap * the pages and send SIGBUS to the processes if the data was dirty. @@ -927,7 +935,6 @@ static int hwpoison_user_mappings(struct page *p, unsigned long pfn, int ret; int kill = 1, forcekill; struct page *hpage = *hpagep; - struct page *ppage; /* * Here we are interested only in user-mapped pages, so skip any @@ -977,59 +984,6 @@ static int hwpoison_user_mappings(struct page *p, unsigned long pfn, } /* - * ppage: poisoned page - * if p is regular page(4k page) - * ppage == real poisoned page; - * else p is hugetlb or THP, ppage == head page. - */ - ppage = hpage; - - if (PageTransHuge(hpage)) { - /* - * Verify that this isn't a hugetlbfs head page, the check for - * PageAnon is just for avoid tripping a split_huge_page - * internal debug check, as split_huge_page refuses to deal with - * anything that isn't an anon page. PageAnon can't go away fro - * under us because we hold a refcount on the hpage, without a - * refcount on the hpage. split_huge_page can't be safely called - * in the first place, having a refcount on the tail isn't - * enough * to be safe. - */ - if (!PageHuge(hpage) && PageAnon(hpage)) { - if (unlikely(split_huge_page(hpage))) { - /* - * FIXME: if splitting THP is failed, it is - * better to stop the following operation rather - * than causing panic by unmapping. System might - * survive if the page is freed later. - */ - printk(KERN_INFO - "MCE %#lx: failed to split THP\n", pfn); - - BUG_ON(!PageHWPoison(p)); - return SWAP_FAIL; - } - /* - * We pinned the head page for hwpoison handling, - * now we split the thp and we are interested in - * the hwpoisoned raw page, so move the refcount - * to it. Similarly, page lock is shifted. - */ - if (hpage != p) { - if (!(flags & MF_COUNT_INCREASED)) { - put_page(hpage); - get_page(p); - } - lock_page(p); - unlock_page(hpage); - *hpagep = p; - } - /* THP is split, so ppage should be the real poisoned page. */ - ppage = p; - } - } - - /* * First collect all the processes that have the page * mapped in dirty form. This has to be done before try_to_unmap, * because ttu takes the rmap data structures down. @@ -1038,12 +992,12 @@ static int hwpoison_user_mappings(struct page *p, unsigned long pfn, * there's nothing that can be done. */ if (kill) - collect_procs(ppage, &tokill, flags & MF_ACTION_REQUIRED); + collect_procs(hpage, &tokill, flags & MF_ACTION_REQUIRED); - ret = try_to_unmap(ppage, ttu); + ret = try_to_unmap(hpage, ttu); if (ret != SWAP_SUCCESS) printk(KERN_ERR "MCE %#lx: failed to unmap page (mapcount=%d)\n", - pfn, page_mapcount(ppage)); + pfn, page_mapcount(hpage)); /* * Now that the dirty bit has been propagated to the @@ -1055,7 +1009,7 @@ static int hwpoison_user_mappings(struct page *p, unsigned long pfn, * use a more force-full uncatchable kill to prevent * any accesses to the poisoned memory. */ - forcekill = PageDirty(ppage) || (flags & MF_MUST_KILL); + forcekill = PageDirty(hpage) || (flags & MF_MUST_KILL); kill_procs(&tokill, forcekill, trapno, ret != SWAP_SUCCESS, p, pfn, flags); @@ -1101,6 +1055,7 @@ int memory_failure(unsigned long pfn, int trapno, int flags) struct page_state *ps; struct page *p; struct page *hpage; + struct page *orig_head; int res; unsigned int nr_pages; unsigned long page_flags; @@ -1116,7 +1071,7 @@ int memory_failure(unsigned long pfn, int trapno, int flags) } p = pfn_to_page(pfn); - hpage = compound_head(p); + orig_head = hpage = compound_head(p); if (TestSetPageHWPoison(p)) { printk(KERN_ERR "MCE %#lx: already hardware poisoned\n", pfn); return 0; @@ -1149,10 +1104,9 @@ int memory_failure(unsigned long pfn, int trapno, int flags) * In fact it's dangerous to directly bump up page count from 0, * that may make page_freeze_refs()/page_unfreeze_refs() mismatch. */ - if (!(flags & MF_COUNT_INCREASED) && - !get_page_unless_zero(hpage)) { + if (!(flags & MF_COUNT_INCREASED) && !get_hwpoison_page(p)) { if (is_free_buddy_page(p)) { - action_result(pfn, MSG_BUDDY, DELAYED); + action_result(pfn, MF_MSG_BUDDY, MF_DELAYED); return 0; } else if (PageHuge(hpage)) { /* @@ -1169,16 +1123,39 @@ int memory_failure(unsigned long pfn, int trapno, int flags) } set_page_hwpoison_huge_page(hpage); res = dequeue_hwpoisoned_huge_page(hpage); - action_result(pfn, MSG_FREE_HUGE, - res ? IGNORED : DELAYED); + action_result(pfn, MF_MSG_FREE_HUGE, + res ? MF_IGNORED : MF_DELAYED); unlock_page(hpage); return res; } else { - action_result(pfn, MSG_KERNEL_HIGH_ORDER, IGNORED); + action_result(pfn, MF_MSG_KERNEL_HIGH_ORDER, MF_IGNORED); return -EBUSY; } } + if (!PageHuge(p) && PageTransHuge(hpage)) { + if (!PageAnon(hpage)) { + pr_err("MCE: %#lx: non anonymous thp\n", pfn); + if (TestClearPageHWPoison(p)) + atomic_long_sub(nr_pages, &num_poisoned_pages); + put_page(p); + if (p != hpage) + put_page(hpage); + return -EBUSY; + } + if (unlikely(split_huge_page(hpage))) { + pr_err("MCE: %#lx: thp split failed\n", pfn); + if (TestClearPageHWPoison(p)) + atomic_long_sub(nr_pages, &num_poisoned_pages); + put_page(p); + if (p != hpage) + put_page(hpage); + return -EBUSY; + } + VM_BUG_ON_PAGE(!page_count(p), p); + hpage = compound_head(p); + } + /* * We ignore non-LRU pages for good reasons. * - PG_locked is only well defined for LRU pages and a few others @@ -1188,18 +1165,18 @@ int memory_failure(unsigned long pfn, int trapno, int flags) * walked by the page reclaim code, however that's not a big loss. */ if (!PageHuge(p)) { - if (!PageLRU(hpage)) - shake_page(hpage, 0); - if (!PageLRU(hpage)) { + if (!PageLRU(p)) + shake_page(p, 0); + if (!PageLRU(p)) { /* * shake_page could have turned it free. */ if (is_free_buddy_page(p)) { if (flags & MF_COUNT_INCREASED) - action_result(pfn, MSG_BUDDY, DELAYED); + action_result(pfn, MF_MSG_BUDDY, MF_DELAYED); else - action_result(pfn, MSG_BUDDY_2ND, - DELAYED); + action_result(pfn, MF_MSG_BUDDY_2ND, + MF_DELAYED); return 0; } } @@ -1211,8 +1188,8 @@ int memory_failure(unsigned long pfn, int trapno, int flags) * The page could have changed compound pages during the locking. * If this happens just bail out. */ - if (compound_head(p) != hpage) { - action_result(pfn, MSG_DIFFERENT_COMPOUND, IGNORED); + if (PageCompound(p) && compound_head(p) != orig_head) { + action_result(pfn, MF_MSG_DIFFERENT_COMPOUND, MF_IGNORED); res = -EBUSY; goto out; } @@ -1252,7 +1229,7 @@ int memory_failure(unsigned long pfn, int trapno, int flags) * on the head page to show that the hugepage is hwpoisoned */ if (PageHuge(p) && PageTail(p) && TestSetPageHWPoison(hpage)) { - action_result(pfn, MSG_POISONED_HUGE, IGNORED); + action_result(pfn, MF_MSG_POISONED_HUGE, MF_IGNORED); unlock_page(hpage); put_page(hpage); return 0; @@ -1281,7 +1258,7 @@ int memory_failure(unsigned long pfn, int trapno, int flags) */ if (hwpoison_user_mappings(p, pfn, trapno, flags, &hpage) != SWAP_SUCCESS) { - action_result(pfn, MSG_UNMAP_FAILED, IGNORED); + action_result(pfn, MF_MSG_UNMAP_FAILED, MF_IGNORED); res = -EBUSY; goto out; } @@ -1290,7 +1267,7 @@ int memory_failure(unsigned long pfn, int trapno, int flags) * Torn down by someone else? */ if (PageLRU(p) && !PageSwapCache(p) && p->mapping == NULL) { - action_result(pfn, MSG_TRUNCATED_LRU, IGNORED); + action_result(pfn, MF_MSG_TRUNCATED_LRU, MF_IGNORED); res = -EBUSY; goto out; } @@ -1450,12 +1427,12 @@ int unpoison_memory(unsigned long pfn) */ if (!PageHuge(page) && PageTransHuge(page)) { pr_info("MCE: Memory failure is now running on %#lx\n", pfn); - return 0; + return 0; } nr_pages = 1 << compound_order(page); - if (!get_page_unless_zero(page)) { + if (!get_hwpoison_page(p)) { /* * Since HWPoisoned hugepage should have non-zero refcount, * race between memory failure and unpoison seems to happen. @@ -1523,7 +1500,7 @@ static int __get_any_page(struct page *p, unsigned long pfn, int flags) * When the target page is a free hugepage, just remove it * from free hugepage list. */ - if (!get_page_unless_zero(compound_head(p))) { + if (!get_hwpoison_page(p)) { if (PageHuge(p)) { pr_info("%s: %#lx free huge page\n", __func__, pfn); ret = 0; @@ -1694,20 +1671,7 @@ static int __soft_offline_page(struct page *page, int flags) if (ret > 0) ret = -EIO; } else { - /* - * After page migration succeeds, the source page can - * be trapped in pagevec and actual freeing is delayed. - * Freeing code works differently based on PG_hwpoison, - * so there's a race. We need to make sure that the - * source page should be freed back to buddy before - * setting PG_hwpoison. - */ - if (!is_free_buddy_page(page)) - drain_all_pages(page_zone(page)); SetPageHWPoison(page); - if (!is_free_buddy_page(page)) - pr_info("soft offline: %#lx: page leaked\n", - pfn); atomic_long_inc(&num_poisoned_pages); } } else { @@ -1759,14 +1723,6 @@ int soft_offline_page(struct page *page, int flags) get_online_mems(); - /* - * Isolate the page, so that it doesn't get reallocated if it - * was free. This flag should be kept set until the source page - * is freed and PG_hwpoison on it is set. - */ - if (get_pageblock_migratetype(page) != MIGRATE_ISOLATE) - set_migratetype_isolate(page, true); - ret = get_any_page(page, pfn, flags); put_online_mems(); if (ret > 0) { /* for in-use pages */ @@ -1785,6 +1741,5 @@ int soft_offline_page(struct page *page, int flags) atomic_long_inc(&num_poisoned_pages); } } - unset_migratetype_isolate(page, MIGRATE_MOVABLE); return ret; } diff --git a/mm/memory.c b/mm/memory.c index 17734c3c1183..11b9ca176740 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -2081,11 +2081,12 @@ static int wp_page_copy(struct mm_struct *mm, struct vm_area_struct *vma, goto oom; cow_user_page(new_page, old_page, address, vma); } - __SetPageUptodate(new_page); if (mem_cgroup_try_charge(new_page, mm, GFP_KERNEL, &memcg)) goto oom_free_new; + __SetPageUptodate(new_page); + mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end); /* @@ -2689,6 +2690,10 @@ static int do_anonymous_page(struct mm_struct *mm, struct vm_area_struct *vma, page = alloc_zeroed_user_highpage_movable(vma, address); if (!page) goto oom; + + if (mem_cgroup_try_charge(page, mm, GFP_KERNEL, &memcg)) + goto oom_free_page; + /* * The memory barrier inside __SetPageUptodate makes sure that * preceeding stores to the page contents become visible before @@ -2696,9 +2701,6 @@ static int do_anonymous_page(struct mm_struct *mm, struct vm_area_struct *vma, */ __SetPageUptodate(page); - if (mem_cgroup_try_charge(page, mm, GFP_KERNEL, &memcg)) - goto oom_free_page; - entry = mk_pte(page, vma->vm_page_prot); if (vma->vm_flags & VM_WRITE) entry = pte_mkwrite(pte_mkdirty(entry)); diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c index 9e88f749aa51..26fbba7d888f 100644 --- a/mm/memory_hotplug.c +++ b/mm/memory_hotplug.c @@ -513,6 +513,7 @@ int __ref __add_pages(int nid, struct zone *zone, unsigned long phys_start_pfn, break; err = 0; } + vmemmap_populate_print_last(); return err; } diff --git a/mm/mempolicy.c b/mm/mempolicy.c index 747743237d9f..99d4c1d0b858 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -1972,35 +1972,41 @@ retry_cpuset: pol = get_vma_policy(vma, addr); cpuset_mems_cookie = read_mems_allowed_begin(); - if (unlikely(IS_ENABLED(CONFIG_TRANSPARENT_HUGEPAGE) && hugepage && - pol->mode != MPOL_INTERLEAVE)) { + if (pol->mode == MPOL_INTERLEAVE) { + unsigned nid; + + nid = interleave_nid(pol, vma, addr, PAGE_SHIFT + order); + mpol_cond_put(pol); + page = alloc_page_interleave(gfp, order, nid); + goto out; + } + + if (unlikely(IS_ENABLED(CONFIG_TRANSPARENT_HUGEPAGE) && hugepage)) { + int hpage_node = node; + /* * For hugepage allocation and non-interleave policy which - * allows the current node, we only try to allocate from the - * current node and don't fall back to other nodes, as the - * cost of remote accesses would likely offset THP benefits. + * allows the current node (or other explicitly preferred + * node) we only try to allocate from the current/preferred + * node and don't fall back to other nodes, as the cost of + * remote accesses would likely offset THP benefits. * * If the policy is interleave, or does not allow the current * node in its nodemask, we allocate the standard way. */ + if (pol->mode == MPOL_PREFERRED && + !(pol->flags & MPOL_F_LOCAL)) + hpage_node = pol->v.preferred_node; + nmask = policy_nodemask(gfp, pol); - if (!nmask || node_isset(node, *nmask)) { + if (!nmask || node_isset(hpage_node, *nmask)) { mpol_cond_put(pol); - page = alloc_pages_exact_node(node, + page = alloc_pages_exact_node(hpage_node, gfp | __GFP_THISNODE, order); goto out; } } - if (pol->mode == MPOL_INTERLEAVE) { - unsigned nid; - - nid = interleave_nid(pol, vma, addr, PAGE_SHIFT + order); - mpol_cond_put(pol); - page = alloc_page_interleave(gfp, order, nid); - goto out; - } - nmask = policy_nodemask(gfp, pol); zl = policy_zonelist(gfp, pol, node); mpol_cond_put(pol); diff --git a/mm/memtest.c b/mm/memtest.c index 1997d934b13b..0a1cc133f6d7 100644 --- a/mm/memtest.c +++ b/mm/memtest.c @@ -74,7 +74,8 @@ static void __init do_one_pass(u64 pattern, phys_addr_t start, phys_addr_t end) u64 i; phys_addr_t this_start, this_end; - for_each_free_mem_range(i, NUMA_NO_NODE, &this_start, &this_end, NULL) { + for_each_free_mem_range(i, NUMA_NO_NODE, MEMBLOCK_NONE, &this_start, + &this_end, NULL) { this_start = clamp(this_start, start, end); this_end = clamp(this_end, start, end); if (this_start < this_end) { diff --git a/mm/migrate.c b/mm/migrate.c index f53838fe3dfe..ee401e4e5ef1 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -918,7 +918,8 @@ out: static ICE_noinline int unmap_and_move(new_page_t get_new_page, free_page_t put_new_page, unsigned long private, struct page *page, - int force, enum migrate_mode mode) + int force, enum migrate_mode mode, + enum migrate_reason reason) { int rc = 0; int *result = NULL; @@ -949,7 +950,8 @@ out: list_del(&page->lru); dec_zone_page_state(page, NR_ISOLATED_ANON + page_is_file_cache(page)); - putback_lru_page(page); + if (reason != MR_MEMORY_FAILURE) + putback_lru_page(page); } /* @@ -1122,7 +1124,8 @@ int migrate_pages(struct list_head *from, new_page_t get_new_page, pass > 2, mode); else rc = unmap_and_move(get_new_page, put_new_page, - private, page, pass > 2, mode); + private, page, pass > 2, mode, + reason); switch(rc) { case -ENOMEM: @@ -1796,7 +1799,7 @@ fail_putback: */ flush_cache_range(vma, mmun_start, mmun_end); page_add_anon_rmap(new_page, vma, mmun_start); - pmdp_clear_flush_notify(vma, mmun_start, pmd); + pmdp_huge_clear_flush_notify(vma, mmun_start, pmd); set_pmd_at(mm, mmun_start, pmd, entry); flush_tlb_range(vma, mmun_start, mmun_end); update_mmu_cache_pmd(vma, address, &entry); diff --git a/mm/mmap.c b/mm/mmap.c index bb50cacc3ea5..aa632ade2be7 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -1258,6 +1258,9 @@ unsigned long do_mmap_pgoff(struct file *file, unsigned long addr, *populate = 0; + if (!len) + return -EINVAL; + /* * Does the application expect PROT_READ to imply PROT_EXEC? * @@ -1268,9 +1271,6 @@ unsigned long do_mmap_pgoff(struct file *file, unsigned long addr, if (!(file && (file->f_path.mnt->mnt_flags & MNT_NOEXEC))) prot |= PROT_EXEC; - if (!len) - return -EINVAL; - if (!(flags & MAP_FIXED)) addr = round_hint_to_min(addr); diff --git a/mm/mprotect.c b/mm/mprotect.c index 88584838e704..e7d6f1171ecb 100644 --- a/mm/mprotect.c +++ b/mm/mprotect.c @@ -29,6 +29,8 @@ #include <asm/cacheflush.h> #include <asm/tlbflush.h> +#include "internal.h" + /* * For a prot_numa update we only hold mmap_sem for read so there is a * potential race with faulting where a pmd was temporarily none. This @@ -322,6 +324,15 @@ success: change_protection(vma, start, end, vma->vm_page_prot, dirty_accountable, 0); + /* + * Private VM_LOCKED VMA becoming writable: trigger COW to avoid major + * fault on access. + */ + if ((oldflags & (VM_WRITE | VM_SHARED | VM_LOCKED)) == VM_LOCKED && + (newflags & VM_WRITE)) { + populate_vma_page_range(vma, start, end, NULL); + } + vm_stat_account(mm, oldflags, vma->vm_file, -nrpages); vm_stat_account(mm, newflags, vma->vm_file, nrpages); perf_event_mmap(vma); diff --git a/mm/mremap.c b/mm/mremap.c index 034e2d360652..a7c93eceb1c8 100644 --- a/mm/mremap.c +++ b/mm/mremap.c @@ -22,6 +22,7 @@ #include <linux/mmu_notifier.h> #include <linux/sched/sysctl.h> #include <linux/uaccess.h> +#include <linux/mm-arch-hooks.h> #include <asm/cacheflush.h> #include <asm/tlbflush.h> @@ -286,13 +287,17 @@ static unsigned long move_vma(struct vm_area_struct *vma, old_len = new_len; old_addr = new_addr; new_addr = -ENOMEM; - } else if (vma->vm_file && vma->vm_file->f_op->mremap) { - err = vma->vm_file->f_op->mremap(vma->vm_file, new_vma); - if (err < 0) { - move_page_tables(new_vma, new_addr, vma, old_addr, - moved_len, true); - return err; + } else { + if (vma->vm_file && vma->vm_file->f_op->mremap) { + err = vma->vm_file->f_op->mremap(vma->vm_file, new_vma); + if (err < 0) { + move_page_tables(new_vma, new_addr, vma, + old_addr, moved_len, true); + return err; + } } + arch_remap(mm, old_addr, old_addr + old_len, + new_addr, new_addr + new_len); } /* Conceal VM_ACCOUNT so old reservation is not undone */ diff --git a/mm/nobootmem.c b/mm/nobootmem.c index 90b50468333e..5258386fa1be 100644 --- a/mm/nobootmem.c +++ b/mm/nobootmem.c @@ -37,11 +37,20 @@ static void * __init __alloc_memory_core_early(int nid, u64 size, u64 align, { void *ptr; u64 addr; + ulong flags = choose_memblock_flags(); if (limit > memblock.current_limit) limit = memblock.current_limit; - addr = memblock_find_in_range_node(size, align, goal, limit, nid); +again: + addr = memblock_find_in_range_node(size, align, goal, limit, nid, + flags); + if (!addr && (flags & MEMBLOCK_MIRROR)) { + flags &= ~MEMBLOCK_MIRROR; + pr_warn("Could not allocate %pap bytes of mirrored memory\n", + &size); + goto again; + } if (!addr) return NULL; @@ -121,7 +130,8 @@ static unsigned long __init free_low_memory_core_early(void) memblock_clear_hotplug(0, -1); - for_each_free_mem_range(i, NUMA_NO_NODE, &start, &end, NULL) + for_each_free_mem_range(i, NUMA_NO_NODE, MEMBLOCK_NONE, &start, &end, + NULL) count += __free_memory_core(start, end); #ifdef CONFIG_ARCH_DISCARD_MEMBLOCK diff --git a/mm/nommu.c b/mm/nommu.c index e544508e2a4b..05e7447d960b 100644 --- a/mm/nommu.c +++ b/mm/nommu.c @@ -42,22 +42,6 @@ #include <asm/mmu_context.h> #include "internal.h" -#if 0 -#define kenter(FMT, ...) \ - printk(KERN_DEBUG "==> %s("FMT")\n", __func__, ##__VA_ARGS__) -#define kleave(FMT, ...) \ - printk(KERN_DEBUG "<== %s()"FMT"\n", __func__, ##__VA_ARGS__) -#define kdebug(FMT, ...) \ - printk(KERN_DEBUG "xxx" FMT"yyy\n", ##__VA_ARGS__) -#else -#define kenter(FMT, ...) \ - no_printk(KERN_DEBUG "==> %s("FMT")\n", __func__, ##__VA_ARGS__) -#define kleave(FMT, ...) \ - no_printk(KERN_DEBUG "<== %s()"FMT"\n", __func__, ##__VA_ARGS__) -#define kdebug(FMT, ...) \ - no_printk(KERN_DEBUG FMT"\n", ##__VA_ARGS__) -#endif - void *high_memory; EXPORT_SYMBOL(high_memory); struct page *mem_map; @@ -665,11 +649,7 @@ static void free_page_series(unsigned long from, unsigned long to) for (; from < to; from += PAGE_SIZE) { struct page *page = virt_to_page(from); - kdebug("- free %lx", from); atomic_long_dec(&mmap_pages_allocated); - if (page_count(page) != 1) - kdebug("free page %p: refcount not one: %d", - page, page_count(page)); put_page(page); } } @@ -683,8 +663,6 @@ static void free_page_series(unsigned long from, unsigned long to) static void __put_nommu_region(struct vm_region *region) __releases(nommu_region_sem) { - kenter("%p{%d}", region, region->vm_usage); - BUG_ON(!nommu_region_tree.rb_node); if (--region->vm_usage == 0) { @@ -697,10 +675,8 @@ static void __put_nommu_region(struct vm_region *region) /* IO memory and memory shared directly out of the pagecache * from ramfs/tmpfs mustn't be released here */ - if (region->vm_flags & VM_MAPPED_COPY) { - kdebug("free series"); + if (region->vm_flags & VM_MAPPED_COPY) free_page_series(region->vm_start, region->vm_top); - } kmem_cache_free(vm_region_jar, region); } else { up_write(&nommu_region_sem); @@ -744,8 +720,6 @@ static void add_vma_to_mm(struct mm_struct *mm, struct vm_area_struct *vma) struct address_space *mapping; struct rb_node **p, *parent, *rb_prev; - kenter(",%p", vma); - BUG_ON(!vma->vm_region); mm->map_count++; @@ -813,8 +787,6 @@ static void delete_vma_from_mm(struct vm_area_struct *vma) struct mm_struct *mm = vma->vm_mm; struct task_struct *curr = current; - kenter("%p", vma); - protect_vma(vma, 0); mm->map_count--; @@ -854,7 +826,6 @@ static void delete_vma_from_mm(struct vm_area_struct *vma) */ static void delete_vma(struct mm_struct *mm, struct vm_area_struct *vma) { - kenter("%p", vma); if (vma->vm_ops && vma->vm_ops->close) vma->vm_ops->close(vma); if (vma->vm_file) @@ -957,12 +928,8 @@ static int validate_mmap_request(struct file *file, int ret; /* do the simple checks first */ - if (flags & MAP_FIXED) { - printk(KERN_DEBUG - "%d: Can't do fixed-address/overlay mmap of RAM\n", - current->pid); + if (flags & MAP_FIXED) return -EINVAL; - } if ((flags & MAP_TYPE) != MAP_PRIVATE && (flags & MAP_TYPE) != MAP_SHARED) @@ -1060,8 +1027,7 @@ static int validate_mmap_request(struct file *file, ) { capabilities &= ~NOMMU_MAP_DIRECT; if (flags & MAP_SHARED) { - printk(KERN_WARNING - "MAP_SHARED not completely supported on !MMU\n"); + pr_warn("MAP_SHARED not completely supported on !MMU\n"); return -EINVAL; } } @@ -1205,16 +1171,12 @@ static int do_mmap_private(struct vm_area_struct *vma, * we're allocating is smaller than a page */ order = get_order(len); - kdebug("alloc order %d for %lx", order, len); - total = 1 << order; point = len >> PAGE_SHIFT; /* we don't want to allocate a power-of-2 sized page set */ - if (sysctl_nr_trim_pages && total - point >= sysctl_nr_trim_pages) { + if (sysctl_nr_trim_pages && total - point >= sysctl_nr_trim_pages) total = point; - kdebug("try to alloc exact %lu pages", total); - } base = alloc_pages_exact(total << PAGE_SHIFT, GFP_KERNEL); if (!base) @@ -1285,18 +1247,14 @@ unsigned long do_mmap_pgoff(struct file *file, unsigned long capabilities, vm_flags, result; int ret; - kenter(",%lx,%lx,%lx,%lx,%lx", addr, len, prot, flags, pgoff); - *populate = 0; /* decide whether we should attempt the mapping, and if so what sort of * mapping */ ret = validate_mmap_request(file, addr, len, prot, flags, pgoff, &capabilities); - if (ret < 0) { - kleave(" = %d [val]", ret); + if (ret < 0) return ret; - } /* we ignore the address hint */ addr = 0; @@ -1383,11 +1341,9 @@ unsigned long do_mmap_pgoff(struct file *file, vma->vm_start = start; vma->vm_end = start + len; - if (pregion->vm_flags & VM_MAPPED_COPY) { - kdebug("share copy"); + if (pregion->vm_flags & VM_MAPPED_COPY) vma->vm_flags |= VM_MAPPED_COPY; - } else { - kdebug("share mmap"); + else { ret = do_mmap_shared_file(vma); if (ret < 0) { vma->vm_region = NULL; @@ -1467,7 +1423,6 @@ share: up_write(&nommu_region_sem); - kleave(" = %lx", result); return result; error_just_free: @@ -1479,27 +1434,24 @@ error: if (vma->vm_file) fput(vma->vm_file); kmem_cache_free(vm_area_cachep, vma); - kleave(" = %d", ret); return ret; sharing_violation: up_write(&nommu_region_sem); - printk(KERN_WARNING "Attempt to share mismatched mappings\n"); + pr_warn("Attempt to share mismatched mappings\n"); ret = -EINVAL; goto error; error_getting_vma: kmem_cache_free(vm_region_jar, region); - printk(KERN_WARNING "Allocation of vma for %lu byte allocation" - " from process %d failed\n", - len, current->pid); + pr_warn("Allocation of vma for %lu byte allocation from process %d failed\n", + len, current->pid); show_free_areas(0); return -ENOMEM; error_getting_region: - printk(KERN_WARNING "Allocation of vm region for %lu byte allocation" - " from process %d failed\n", - len, current->pid); + pr_warn("Allocation of vm region for %lu byte allocation from process %d failed\n", + len, current->pid); show_free_areas(0); return -ENOMEM; } @@ -1563,8 +1515,6 @@ int split_vma(struct mm_struct *mm, struct vm_area_struct *vma, struct vm_region *region; unsigned long npages; - kenter(""); - /* we're only permitted to split anonymous regions (these should have * only a single usage on the region) */ if (vma->vm_file) @@ -1628,8 +1578,6 @@ static int shrink_vma(struct mm_struct *mm, { struct vm_region *region; - kenter(""); - /* adjust the VMA's pointers, which may reposition it in the MM's tree * and list */ delete_vma_from_mm(vma); @@ -1669,8 +1617,6 @@ int do_munmap(struct mm_struct *mm, unsigned long start, size_t len) unsigned long end; int ret; - kenter(",%lx,%zx", start, len); - len = PAGE_ALIGN(len); if (len == 0) return -EINVAL; @@ -1682,11 +1628,9 @@ int do_munmap(struct mm_struct *mm, unsigned long start, size_t len) if (!vma) { static int limit; if (limit < 5) { - printk(KERN_WARNING - "munmap of memory not mmapped by process %d" - " (%s): 0x%lx-0x%lx\n", - current->pid, current->comm, - start, start + len - 1); + pr_warn("munmap of memory not mmapped by process %d (%s): 0x%lx-0x%lx\n", + current->pid, current->comm, + start, start + len - 1); limit++; } return -EINVAL; @@ -1695,38 +1639,27 @@ int do_munmap(struct mm_struct *mm, unsigned long start, size_t len) /* we're allowed to split an anonymous VMA but not a file-backed one */ if (vma->vm_file) { do { - if (start > vma->vm_start) { - kleave(" = -EINVAL [miss]"); + if (start > vma->vm_start) return -EINVAL; - } if (end == vma->vm_end) goto erase_whole_vma; vma = vma->vm_next; } while (vma); - kleave(" = -EINVAL [split file]"); return -EINVAL; } else { /* the chunk must be a subset of the VMA found */ if (start == vma->vm_start && end == vma->vm_end) goto erase_whole_vma; - if (start < vma->vm_start || end > vma->vm_end) { - kleave(" = -EINVAL [superset]"); + if (start < vma->vm_start || end > vma->vm_end) return -EINVAL; - } - if (start & ~PAGE_MASK) { - kleave(" = -EINVAL [unaligned start]"); + if (start & ~PAGE_MASK) return -EINVAL; - } - if (end != vma->vm_end && end & ~PAGE_MASK) { - kleave(" = -EINVAL [unaligned split]"); + if (end != vma->vm_end && end & ~PAGE_MASK) return -EINVAL; - } if (start != vma->vm_start && end != vma->vm_end) { ret = split_vma(mm, vma, start, 1); - if (ret < 0) { - kleave(" = %d [split]", ret); + if (ret < 0) return ret; - } } return shrink_vma(mm, vma, start, end); } @@ -1734,7 +1667,6 @@ int do_munmap(struct mm_struct *mm, unsigned long start, size_t len) erase_whole_vma: delete_vma_from_mm(vma); delete_vma(mm, vma); - kleave(" = 0"); return 0; } EXPORT_SYMBOL(do_munmap); @@ -1766,8 +1698,6 @@ void exit_mmap(struct mm_struct *mm) if (!mm) return; - kenter(""); - mm->total_vm = 0; while ((vma = mm->mmap)) { @@ -1776,8 +1706,6 @@ void exit_mmap(struct mm_struct *mm) delete_vma(mm, vma); cond_resched(); } - - kleave(""); } unsigned long vm_brk(unsigned long addr, unsigned long len) diff --git a/mm/oom_kill.c b/mm/oom_kill.c index 2b665da1b3c9..dff991e0681e 100644 --- a/mm/oom_kill.c +++ b/mm/oom_kill.c @@ -42,7 +42,8 @@ int sysctl_panic_on_oom; int sysctl_oom_kill_allocating_task; int sysctl_oom_dump_tasks = 1; -static DEFINE_SPINLOCK(zone_scan_lock); + +DEFINE_MUTEX(oom_lock); #ifdef CONFIG_NUMA /** @@ -405,16 +406,15 @@ static atomic_t oom_victims = ATOMIC_INIT(0); static DECLARE_WAIT_QUEUE_HEAD(oom_victims_wait); bool oom_killer_disabled __read_mostly; -static DECLARE_RWSEM(oom_sem); /** - * mark_tsk_oom_victim - marks the given task as OOM victim. + * mark_oom_victim - mark the given task as OOM victim * @tsk: task to mark * - * Has to be called with oom_sem taken for read and never after + * Has to be called with oom_lock held and never after * oom has been disabled already. */ -void mark_tsk_oom_victim(struct task_struct *tsk) +void mark_oom_victim(struct task_struct *tsk) { WARN_ON(oom_killer_disabled); /* OOM killer might race with memcg OOM */ @@ -431,23 +431,14 @@ void mark_tsk_oom_victim(struct task_struct *tsk) } /** - * unmark_oom_victim - unmarks the current task as OOM victim. - * - * Wakes up all waiters in oom_killer_disable() + * exit_oom_victim - note the exit of an OOM victim */ -void unmark_oom_victim(void) +void exit_oom_victim(void) { - if (!test_and_clear_thread_flag(TIF_MEMDIE)) - return; + clear_thread_flag(TIF_MEMDIE); - down_read(&oom_sem); - /* - * There is no need to signal the lasst oom_victim if there - * is nobody who cares. - */ - if (!atomic_dec_return(&oom_victims) && oom_killer_disabled) + if (!atomic_dec_return(&oom_victims)) wake_up_all(&oom_victims_wait); - up_read(&oom_sem); } /** @@ -469,14 +460,14 @@ bool oom_killer_disable(void) * Make sure to not race with an ongoing OOM killer * and that the current is not the victim. */ - down_write(&oom_sem); + mutex_lock(&oom_lock); if (test_thread_flag(TIF_MEMDIE)) { - up_write(&oom_sem); + mutex_unlock(&oom_lock); return false; } oom_killer_disabled = true; - up_write(&oom_sem); + mutex_unlock(&oom_lock); wait_event(oom_victims_wait, !atomic_read(&oom_victims)); @@ -488,9 +479,7 @@ bool oom_killer_disable(void) */ void oom_killer_enable(void) { - down_write(&oom_sem); oom_killer_disabled = false; - up_write(&oom_sem); } #define K(x) ((x) << (PAGE_SHIFT-10)) @@ -517,7 +506,7 @@ void oom_kill_process(struct task_struct *p, gfp_t gfp_mask, int order, */ task_lock(p); if (p->mm && task_will_free_mem(p)) { - mark_tsk_oom_victim(p); + mark_oom_victim(p); task_unlock(p); put_task_struct(p); return; @@ -528,7 +517,7 @@ void oom_kill_process(struct task_struct *p, gfp_t gfp_mask, int order, dump_header(p, gfp_mask, order, memcg, nodemask); task_lock(p); - pr_err("%s: Kill process %d (%s) score %d or sacrifice child\n", + pr_err("%s: Kill process %d (%s) score %u or sacrifice child\n", message, task_pid_nr(p), p->comm, points); task_unlock(p); @@ -572,7 +561,7 @@ void oom_kill_process(struct task_struct *p, gfp_t gfp_mask, int order, /* mm cannot safely be dereferenced after task_unlock(victim) */ mm = victim->mm; - mark_tsk_oom_victim(victim); + mark_oom_victim(victim); pr_err("Killed process %d (%s) total-vm:%lukB, anon-rss:%lukB, file-rss:%lukB\n", task_pid_nr(victim), victim->comm, K(victim->mm->total_vm), K(get_mm_counter(victim->mm, MM_ANONPAGES)), @@ -645,52 +634,6 @@ int unregister_oom_notifier(struct notifier_block *nb) } EXPORT_SYMBOL_GPL(unregister_oom_notifier); -/* - * Try to acquire the OOM killer lock for the zones in zonelist. Returns zero - * if a parallel OOM killing is already taking place that includes a zone in - * the zonelist. Otherwise, locks all zones in the zonelist and returns 1. - */ -bool oom_zonelist_trylock(struct zonelist *zonelist, gfp_t gfp_mask) -{ - struct zoneref *z; - struct zone *zone; - bool ret = true; - - spin_lock(&zone_scan_lock); - for_each_zone_zonelist(zone, z, zonelist, gfp_zone(gfp_mask)) - if (test_bit(ZONE_OOM_LOCKED, &zone->flags)) { - ret = false; - goto out; - } - - /* - * Lock each zone in the zonelist under zone_scan_lock so a parallel - * call to oom_zonelist_trylock() doesn't succeed when it shouldn't. - */ - for_each_zone_zonelist(zone, z, zonelist, gfp_zone(gfp_mask)) - set_bit(ZONE_OOM_LOCKED, &zone->flags); - -out: - spin_unlock(&zone_scan_lock); - return ret; -} - -/* - * Clears the ZONE_OOM_LOCKED flag for all zones in the zonelist so that failed - * allocation attempts with zonelists containing them may now recall the OOM - * killer, if necessary. - */ -void oom_zonelist_unlock(struct zonelist *zonelist, gfp_t gfp_mask) -{ - struct zoneref *z; - struct zone *zone; - - spin_lock(&zone_scan_lock); - for_each_zone_zonelist(zone, z, zonelist, gfp_zone(gfp_mask)) - clear_bit(ZONE_OOM_LOCKED, &zone->flags); - spin_unlock(&zone_scan_lock); -} - /** * __out_of_memory - kill the "best" process when we run out of memory * @zonelist: zonelist pointer @@ -704,8 +647,8 @@ void oom_zonelist_unlock(struct zonelist *zonelist, gfp_t gfp_mask) * OR try to be smart about which process to kill. Note that we * don't have to be perfect here, we just have to be good. */ -static void __out_of_memory(struct zonelist *zonelist, gfp_t gfp_mask, - int order, nodemask_t *nodemask, bool force_kill) +bool out_of_memory(struct zonelist *zonelist, gfp_t gfp_mask, + int order, nodemask_t *nodemask, bool force_kill) { const nodemask_t *mpol_mask; struct task_struct *p; @@ -715,10 +658,13 @@ static void __out_of_memory(struct zonelist *zonelist, gfp_t gfp_mask, enum oom_constraint constraint = CONSTRAINT_NONE; int killed = 0; + if (oom_killer_disabled) + return false; + blocking_notifier_call_chain(&oom_notify_list, 0, &freed); if (freed > 0) /* Got some memory back in the last second. */ - return; + goto out; /* * If current has a pending SIGKILL or is exiting, then automatically @@ -730,8 +676,8 @@ static void __out_of_memory(struct zonelist *zonelist, gfp_t gfp_mask, */ if (current->mm && (fatal_signal_pending(current) || task_will_free_mem(current))) { - mark_tsk_oom_victim(current); - return; + mark_oom_victim(current); + goto out; } /* @@ -771,32 +717,8 @@ out: */ if (killed) schedule_timeout_killable(1); -} - -/** - * out_of_memory - tries to invoke OOM killer. - * @zonelist: zonelist pointer - * @gfp_mask: memory allocation flags - * @order: amount of memory being requested as a power of 2 - * @nodemask: nodemask passed to page allocator - * @force_kill: true if a task must be killed, even if others are exiting - * - * invokes __out_of_memory if the OOM is not disabled by oom_killer_disable() - * when it returns false. Otherwise returns true. - */ -bool out_of_memory(struct zonelist *zonelist, gfp_t gfp_mask, - int order, nodemask_t *nodemask, bool force_kill) -{ - bool ret = false; - - down_read(&oom_sem); - if (!oom_killer_disabled) { - __out_of_memory(zonelist, gfp_mask, order, nodemask, force_kill); - ret = true; - } - up_read(&oom_sem); - return ret; + return true; } /* @@ -806,27 +728,21 @@ bool out_of_memory(struct zonelist *zonelist, gfp_t gfp_mask, */ void pagefault_out_of_memory(void) { - struct zonelist *zonelist; - - down_read(&oom_sem); if (mem_cgroup_oom_synchronize(true)) - goto unlock; + return; - zonelist = node_zonelist(first_memory_node, GFP_KERNEL); - if (oom_zonelist_trylock(zonelist, GFP_KERNEL)) { - if (!oom_killer_disabled) - __out_of_memory(NULL, 0, 0, NULL, false); - else - /* - * There shouldn't be any user tasks runable while the - * OOM killer is disabled so the current task has to - * be a racing OOM victim for which oom_killer_disable() - * is waiting for. - */ - WARN_ON(test_thread_flag(TIF_MEMDIE)); + if (!mutex_trylock(&oom_lock)) + return; - oom_zonelist_unlock(zonelist, GFP_KERNEL); + if (!out_of_memory(NULL, 0, 0, NULL, false)) { + /* + * There shouldn't be any user tasks runnable while the + * OOM killer is disabled, so the current task has to + * be a racing OOM victim for which oom_killer_disable() + * is waiting for. + */ + WARN_ON(test_thread_flag(TIF_MEMDIE)); } -unlock: - up_read(&oom_sem); + + mutex_unlock(&oom_lock); } diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 2fd31aebef30..5e6fa06f2784 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -380,20 +380,6 @@ void prep_compound_page(struct page *page, unsigned long order) } } -static inline void prep_zero_page(struct page *page, unsigned int order, - gfp_t gfp_flags) -{ - int i; - - /* - * clear_highpage() will use KM_USER0, so it's a bug to use __GFP_ZERO - * and __GFP_HIGHMEM from hard or soft interrupt context. - */ - VM_BUG_ON((gfp_flags & __GFP_HIGHMEM) && in_interrupt()); - for (i = 0; i < (1 << order); i++) - clear_highpage(page + i); -} - #ifdef CONFIG_DEBUG_PAGEALLOC unsigned int _debug_guardpage_minorder; bool _debug_pagealloc_enabled __read_mostly; @@ -975,7 +961,8 @@ static int prep_new_page(struct page *page, unsigned int order, gfp_t gfp_flags, kasan_alloc_pages(page, order); if (gfp_flags & __GFP_ZERO) - prep_zero_page(page, order, gfp_flags); + for (i = 0; i < (1 << order); i++) + clear_highpage(page + i); if (order && (gfp_flags & __GFP_COMP)) prep_compound_page(page, order); @@ -2322,48 +2309,6 @@ void warn_alloc_failed(gfp_t gfp_mask, int order, const char *fmt, ...) show_mem(filter); } -static inline int -should_alloc_retry(gfp_t gfp_mask, unsigned int order, - unsigned long did_some_progress, - unsigned long pages_reclaimed) -{ - /* Do not loop if specifically requested */ - if (gfp_mask & __GFP_NORETRY) - return 0; - - /* Always retry if specifically requested */ - if (gfp_mask & __GFP_NOFAIL) - return 1; - - /* - * Suspend converts GFP_KERNEL to __GFP_WAIT which can prevent reclaim - * making forward progress without invoking OOM. Suspend also disables - * storage devices so kswapd will not help. Bail if we are suspending. - */ - if (!did_some_progress && pm_suspended_storage()) - return 0; - - /* - * In this implementation, order <= PAGE_ALLOC_COSTLY_ORDER - * means __GFP_NOFAIL, but that may not be true in other - * implementations. - */ - if (order <= PAGE_ALLOC_COSTLY_ORDER) - return 1; - - /* - * For order > PAGE_ALLOC_COSTLY_ORDER, if __GFP_REPEAT is - * specified, then we retry until we no longer reclaim any pages - * (above), or we've reclaimed an order of pages at least as - * large as the allocation's order. In both cases, if the - * allocation still fails, we stop retrying. - */ - if (gfp_mask & __GFP_REPEAT && pages_reclaimed < (1 << order)) - return 1; - - return 0; -} - static inline struct page * __alloc_pages_may_oom(gfp_t gfp_mask, unsigned int order, const struct alloc_context *ac, unsigned long *did_some_progress) @@ -2373,10 +2318,10 @@ __alloc_pages_may_oom(gfp_t gfp_mask, unsigned int order, *did_some_progress = 0; /* - * Acquire the per-zone oom lock for each zone. If that - * fails, somebody else is making progress for us. + * Acquire the oom lock. If that fails, somebody else is + * making progress for us. */ - if (!oom_zonelist_trylock(ac->zonelist, gfp_mask)) { + if (!mutex_trylock(&oom_lock)) { *did_some_progress = 1; schedule_timeout_uninterruptible(1); return NULL; @@ -2402,16 +2347,18 @@ __alloc_pages_may_oom(gfp_t gfp_mask, unsigned int order, /* The OOM killer does not needlessly kill tasks for lowmem */ if (ac->high_zoneidx < ZONE_NORMAL) goto out; - /* The OOM killer does not compensate for light reclaim */ + /* The OOM killer does not compensate for IO-less reclaim */ if (!(gfp_mask & __GFP_FS)) { /* * XXX: Page reclaim didn't yield anything, * and the OOM killer can't be invoked, but - * keep looping as per should_alloc_retry(). + * keep looping as per tradition. */ *did_some_progress = 1; goto out; } + if (pm_suspended_storage()) + goto out; /* The OOM killer may not free memory on a specific node */ if (gfp_mask & __GFP_THISNODE) goto out; @@ -2421,7 +2368,7 @@ __alloc_pages_may_oom(gfp_t gfp_mask, unsigned int order, || WARN_ON_ONCE(gfp_mask & __GFP_NOFAIL)) *did_some_progress = 1; out: - oom_zonelist_unlock(ac->zonelist, gfp_mask); + mutex_unlock(&oom_lock); return page; } @@ -2794,40 +2741,40 @@ retry: if (page) goto got_pg; - /* Check if we should retry the allocation */ + /* Do not loop if specifically requested */ + if (gfp_mask & __GFP_NORETRY) + goto noretry; + + /* Keep reclaiming pages as long as there is reasonable progress */ pages_reclaimed += did_some_progress; - if (should_alloc_retry(gfp_mask, order, did_some_progress, - pages_reclaimed)) { - /* - * If we fail to make progress by freeing individual - * pages, but the allocation wants us to keep going, - * start OOM killing tasks. - */ - if (!did_some_progress) { - page = __alloc_pages_may_oom(gfp_mask, order, ac, - &did_some_progress); - if (page) - goto got_pg; - if (!did_some_progress) - goto nopage; - } + if ((did_some_progress && order <= PAGE_ALLOC_COSTLY_ORDER) || + ((gfp_mask & __GFP_REPEAT) && pages_reclaimed < (1 << order))) { /* Wait for some write requests to complete then retry */ wait_iff_congested(ac->preferred_zone, BLK_RW_ASYNC, HZ/50); goto retry; - } else { - /* - * High-order allocations do not necessarily loop after - * direct reclaim and reclaim/compaction depends on compaction - * being called after reclaim so call directly if necessary - */ - page = __alloc_pages_direct_compact(gfp_mask, order, - alloc_flags, ac, migration_mode, - &contended_compaction, - &deferred_compaction); - if (page) - goto got_pg; } + /* Reclaim has failed us, start killing things */ + page = __alloc_pages_may_oom(gfp_mask, order, ac, &did_some_progress); + if (page) + goto got_pg; + + /* Retry as long as the OOM killer is making progress */ + if (did_some_progress) + goto retry; + +noretry: + /* + * High-order allocations do not necessarily loop after + * direct reclaim and reclaim/compaction depends on compaction + * being called after reclaim so call directly if necessary + */ + page = __alloc_pages_direct_compact(gfp_mask, order, alloc_flags, + ac, migration_mode, + &contended_compaction, + &deferred_compaction); + if (page) + goto got_pg; nopage: warn_alloc_failed(gfp_mask, order, NULL); got_pg: @@ -4867,22 +4814,28 @@ static void __meminit calculate_node_totalpages(struct pglist_data *pgdat, unsigned long *zones_size, unsigned long *zholes_size) { - unsigned long realtotalpages, totalpages = 0; + unsigned long realtotalpages = 0, totalpages = 0; enum zone_type i; - for (i = 0; i < MAX_NR_ZONES; i++) - totalpages += zone_spanned_pages_in_node(pgdat->node_id, i, - node_start_pfn, - node_end_pfn, - zones_size); - pgdat->node_spanned_pages = totalpages; - - realtotalpages = totalpages; - for (i = 0; i < MAX_NR_ZONES; i++) - realtotalpages -= - zone_absent_pages_in_node(pgdat->node_id, i, + for (i = 0; i < MAX_NR_ZONES; i++) { + struct zone *zone = pgdat->node_zones + i; + unsigned long size, real_size; + + size = zone_spanned_pages_in_node(pgdat->node_id, i, + node_start_pfn, + node_end_pfn, + zones_size); + real_size = size - zone_absent_pages_in_node(pgdat->node_id, i, node_start_pfn, node_end_pfn, zholes_size); + zone->spanned_pages = size; + zone->present_pages = real_size; + + totalpages += size; + realtotalpages += real_size; + } + + pgdat->node_spanned_pages = totalpages; pgdat->node_present_pages = realtotalpages; printk(KERN_DEBUG "On node %d totalpages: %lu\n", pgdat->node_id, realtotalpages); @@ -4992,8 +4945,7 @@ static unsigned long __paginginit calc_memmap_size(unsigned long spanned_pages, * NOTE: pgdat should get zeroed by caller. */ static void __paginginit free_area_init_core(struct pglist_data *pgdat, - unsigned long node_start_pfn, unsigned long node_end_pfn, - unsigned long *zones_size, unsigned long *zholes_size) + unsigned long node_start_pfn, unsigned long node_end_pfn) { enum zone_type j; int nid = pgdat->node_id; @@ -5014,12 +4966,8 @@ static void __paginginit free_area_init_core(struct pglist_data *pgdat, struct zone *zone = pgdat->node_zones + j; unsigned long size, realsize, freesize, memmap_pages; - size = zone_spanned_pages_in_node(nid, j, node_start_pfn, - node_end_pfn, zones_size); - realsize = freesize = size - zone_absent_pages_in_node(nid, j, - node_start_pfn, - node_end_pfn, - zholes_size); + size = zone->spanned_pages; + realsize = freesize = zone->present_pages; /* * Adjust freesize so that it accounts for how much memory @@ -5054,8 +5002,6 @@ static void __paginginit free_area_init_core(struct pglist_data *pgdat, nr_kernel_pages -= memmap_pages; nr_all_pages += freesize; - zone->spanned_pages = size; - zone->present_pages = realsize; /* * Set an approximate value for lowmem here, it will be adjusted * when the bootmem allocator frees pages into the buddy system. @@ -5161,8 +5107,7 @@ void __paginginit free_area_init_node(int nid, unsigned long *zones_size, (unsigned long)pgdat->node_mem_map); #endif - free_area_init_core(pgdat, start_pfn, end_pfn, - zones_size, zholes_size); + free_area_init_core(pgdat, start_pfn, end_pfn); } #ifdef CONFIG_HAVE_MEMBLOCK_NODE_MAP @@ -6111,9 +6056,9 @@ out: return ret; } +#ifdef CONFIG_NUMA int hashdist = HASHDIST_DEFAULT; -#ifdef CONFIG_NUMA static int __init set_hashdist(char *str) { if (!str) diff --git a/mm/percpu.c b/mm/percpu.c index dfd02484e8de..2dd74487a0af 100644 --- a/mm/percpu.c +++ b/mm/percpu.c @@ -1030,7 +1030,7 @@ area_found: memset((void *)pcpu_chunk_addr(chunk, cpu, 0) + off, 0, size); ptr = __addr_to_pcpu_ptr(chunk->base_addr + off); - kmemleak_alloc_percpu(ptr, size); + kmemleak_alloc_percpu(ptr, size, gfp); return ptr; fail_unlock: diff --git a/mm/pgtable-generic.c b/mm/pgtable-generic.c index c25f94b33811..6b674e00153c 100644 --- a/mm/pgtable-generic.c +++ b/mm/pgtable-generic.c @@ -119,14 +119,15 @@ pte_t ptep_clear_flush(struct vm_area_struct *vma, unsigned long address, } #endif -#ifndef __HAVE_ARCH_PMDP_CLEAR_FLUSH +#ifndef __HAVE_ARCH_PMDP_HUGE_CLEAR_FLUSH #ifdef CONFIG_TRANSPARENT_HUGEPAGE -pmd_t pmdp_clear_flush(struct vm_area_struct *vma, unsigned long address, - pmd_t *pmdp) +pmd_t pmdp_huge_clear_flush(struct vm_area_struct *vma, unsigned long address, + pmd_t *pmdp) { pmd_t pmd; VM_BUG_ON(address & ~HPAGE_PMD_MASK); - pmd = pmdp_get_and_clear(vma->vm_mm, address, pmdp); + VM_BUG_ON(!pmd_trans_huge(*pmdp)); + pmd = pmdp_huge_get_and_clear(vma->vm_mm, address, pmdp); flush_tlb_range(vma, address, address + HPAGE_PMD_SIZE); return pmd; } @@ -198,3 +199,23 @@ void pmdp_invalidate(struct vm_area_struct *vma, unsigned long address, } #endif /* CONFIG_TRANSPARENT_HUGEPAGE */ #endif + +#ifndef pmdp_collapse_flush +#ifdef CONFIG_TRANSPARENT_HUGEPAGE +pmd_t pmdp_collapse_flush(struct vm_area_struct *vma, unsigned long address, + pmd_t *pmdp) +{ + /* + * pmd and hugepage pte format are same. So we could + * use the same function. + */ + pmd_t pmd; + + VM_BUG_ON(address & ~HPAGE_PMD_MASK); + VM_BUG_ON(pmd_trans_huge(*pmdp)); + pmd = pmdp_huge_get_and_clear(vma->vm_mm, address, pmdp); + flush_tlb_range(vma, address, address + HPAGE_PMD_SIZE); + return pmd; +} +#endif /* CONFIG_TRANSPARENT_HUGEPAGE */ +#endif diff --git a/mm/rmap.c b/mm/rmap.c index 24dd3f9fee27..7af1ecb21ccb 100644 --- a/mm/rmap.c +++ b/mm/rmap.c @@ -625,7 +625,7 @@ pmd_t *mm_find_pmd(struct mm_struct *mm, unsigned long address) pmd = pmd_offset(pud, address); /* - * Some THP functions use the sequence pmdp_clear_flush(), set_pmd_at() + * Some THP functions use the sequence pmdp_huge_clear_flush(), set_pmd_at() * without holding anon_vma lock for write. So when looking for a * genuine pmde (in which to find pte), test present and !THP together. */ @@ -950,7 +950,12 @@ void page_move_anon_rmap(struct page *page, VM_BUG_ON_PAGE(page->index != linear_page_index(vma, address), page); anon_vma = (void *) anon_vma + PAGE_MAPPING_ANON; - page->mapping = (struct address_space *) anon_vma; + /* + * Ensure that anon_vma and the PAGE_MAPPING_ANON bit are written + * simultaneously, so a concurrent reader (eg page_referenced()'s + * PageAnon()) will not see one without the other. + */ + WRITE_ONCE(page->mapping, (struct address_space *) anon_vma); } /** diff --git a/mm/shmem.c b/mm/shmem.c index 3759099d8ce4..4caf8ed24d65 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -569,7 +569,7 @@ static int shmem_setattr(struct dentry *dentry, struct iattr *attr) i_size_write(inode, newsize); inode->i_ctime = inode->i_mtime = CURRENT_TIME; } - if (newsize < oldsize) { + if (newsize <= oldsize) { loff_t holebegin = round_up(newsize, PAGE_SIZE); unmap_mapping_range(inode->i_mapping, holebegin, 0, 1); shmem_truncate_range(inode, newsize, (loff_t)-1); diff --git a/mm/slab.c b/mm/slab.c index 7eb38dd1cefa..200e22412a16 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -1454,6 +1454,7 @@ void __init kmem_cache_init(void) kmalloc_caches[INDEX_NODE] = create_kmalloc_cache("kmalloc-node", kmalloc_size(INDEX_NODE), ARCH_KMALLOC_FLAGS); slab_state = PARTIAL_NODE; + setup_kmalloc_cache_index_table(); slab_early_init = 0; diff --git a/mm/slab.h b/mm/slab.h index 4c3ac12dd644..8da63e4e470f 100644 --- a/mm/slab.h +++ b/mm/slab.h @@ -71,6 +71,7 @@ unsigned long calculate_alignment(unsigned long flags, #ifndef CONFIG_SLOB /* Kmalloc array related functions */ +void setup_kmalloc_cache_index_table(void); void create_kmalloc_caches(unsigned long); /* Find the kmalloc slab corresponding for a certain size */ diff --git a/mm/slab_common.c b/mm/slab_common.c index 999bb3424d44..9f8d71f78404 100644 --- a/mm/slab_common.c +++ b/mm/slab_common.c @@ -784,25 +784,45 @@ struct kmem_cache *kmalloc_slab(size_t size, gfp_t flags) } /* - * Create the kmalloc array. Some of the regular kmalloc arrays - * may already have been created because they were needed to - * enable allocations for slab creation. + * kmalloc_info[] is to make slub_debug=,kmalloc-xx option work at boot time. + * kmalloc_index() supports up to 2^26=64MB, so the final entry of the table is + * kmalloc-67108864. */ -void __init create_kmalloc_caches(unsigned long flags) +static struct { + const char *name; + unsigned long size; +} const kmalloc_info[] __initconst = { + {NULL, 0}, {"kmalloc-96", 96}, + {"kmalloc-192", 192}, {"kmalloc-8", 8}, + {"kmalloc-16", 16}, {"kmalloc-32", 32}, + {"kmalloc-64", 64}, {"kmalloc-128", 128}, + {"kmalloc-256", 256}, {"kmalloc-512", 512}, + {"kmalloc-1024", 1024}, {"kmalloc-2048", 2048}, + {"kmalloc-4096", 4096}, {"kmalloc-8192", 8192}, + {"kmalloc-16384", 16384}, {"kmalloc-32768", 32768}, + {"kmalloc-65536", 65536}, {"kmalloc-131072", 131072}, + {"kmalloc-262144", 262144}, {"kmalloc-524288", 524288}, + {"kmalloc-1048576", 1048576}, {"kmalloc-2097152", 2097152}, + {"kmalloc-4194304", 4194304}, {"kmalloc-8388608", 8388608}, + {"kmalloc-16777216", 16777216}, {"kmalloc-33554432", 33554432}, + {"kmalloc-67108864", 67108864} +}; + +/* + * Patch up the size_index table if we have strange large alignment + * requirements for the kmalloc array. This is only the case for + * MIPS it seems. The standard arches will not generate any code here. + * + * Largest permitted alignment is 256 bytes due to the way we + * handle the index determination for the smaller caches. + * + * Make sure that nothing crazy happens if someone starts tinkering + * around with ARCH_KMALLOC_MINALIGN + */ +void __init setup_kmalloc_cache_index_table(void) { int i; - /* - * Patch up the size_index table if we have strange large alignment - * requirements for the kmalloc array. This is only the case for - * MIPS it seems. The standard arches will not generate any code here. - * - * Largest permitted alignment is 256 bytes due to the way we - * handle the index determination for the smaller caches. - * - * Make sure that nothing crazy happens if someone starts tinkering - * around with ARCH_KMALLOC_MINALIGN - */ BUILD_BUG_ON(KMALLOC_MIN_SIZE > 256 || (KMALLOC_MIN_SIZE & (KMALLOC_MIN_SIZE - 1))); @@ -833,39 +853,41 @@ void __init create_kmalloc_caches(unsigned long flags) for (i = 128 + 8; i <= 192; i += 8) size_index[size_index_elem(i)] = 8; } - for (i = KMALLOC_SHIFT_LOW; i <= KMALLOC_SHIFT_HIGH; i++) { +} + +/* + * Create the kmalloc array. Some of the regular kmalloc arrays + * may already have been created because they were needed to + * enable allocations for slab creation. + */ +void __init create_kmalloc_caches(unsigned long flags) +{ + int i; + + for (i = KMALLOC_LOOP_LOW; i <= KMALLOC_SHIFT_HIGH; i++) { if (!kmalloc_caches[i]) { - kmalloc_caches[i] = create_kmalloc_cache(NULL, - 1 << i, flags); + kmalloc_caches[i] = create_kmalloc_cache( + kmalloc_info[i].name, + kmalloc_info[i].size, + flags); } /* - * Caches that are not of the two-to-the-power-of size. - * These have to be created immediately after the - * earlier power of two caches + * "i == 2" is the "kmalloc-192" case which is the last special + * case for initialization and it's the point to jump to + * allocate the minimize size of the object. In slab allocator, + * the KMALLOC_SHIFT_LOW = 5. So, it needs to skip 2^3 and 2^4 + * and go straight to allocate 2^5. If the ARCH_DMA_MINALIGN is + * defined, it may be larger than 2^5 and here is also the + * trick to skip the empty gap. */ - if (KMALLOC_MIN_SIZE <= 32 && !kmalloc_caches[1] && i == 6) - kmalloc_caches[1] = create_kmalloc_cache(NULL, 96, flags); - - if (KMALLOC_MIN_SIZE <= 64 && !kmalloc_caches[2] && i == 7) - kmalloc_caches[2] = create_kmalloc_cache(NULL, 192, flags); + if (i == 2) + i = (KMALLOC_SHIFT_LOW - 1); } /* Kmalloc array is now usable */ slab_state = UP; - for (i = 0; i <= KMALLOC_SHIFT_HIGH; i++) { - struct kmem_cache *s = kmalloc_caches[i]; - char *n; - - if (s) { - n = kasprintf(GFP_NOWAIT, "kmalloc-%d", kmalloc_size(i)); - - BUG_ON(!n); - s->name = n; - } - } - #ifdef CONFIG_ZONE_DMA for (i = 0; i <= KMALLOC_SHIFT_HIGH; i++) { struct kmem_cache *s = kmalloc_caches[i]; diff --git a/mm/slub.c b/mm/slub.c index 54c0876b43d5..816df0016555 100644 --- a/mm/slub.c +++ b/mm/slub.c @@ -3700,6 +3700,7 @@ void __init kmem_cache_init(void) kmem_cache_node = bootstrap(&boot_kmem_cache_node); /* Now we can use the kmem_cache to allocate kmalloc slabs */ + setup_kmalloc_cache_index_table(); create_kmalloc_caches(0); #ifdef CONFIG_SMP diff --git a/mm/swap.c b/mm/swap.c index a7251a8ed532..a3a0a2f1f7c3 100644 --- a/mm/swap.c +++ b/mm/swap.c @@ -131,7 +131,6 @@ void put_unrefcounted_compound_page(struct page *page_head, struct page *page) * here, see the comment above this function. */ VM_BUG_ON_PAGE(!PageHead(page_head), page_head); - VM_BUG_ON_PAGE(page_mapcount(page) != 0, page); if (put_page_testzero(page_head)) { /* * If this is the tail of a slab THP page, diff --git a/mm/vmscan.c b/mm/vmscan.c index 5e8eadd71bac..19ef01e90ac4 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -2646,7 +2646,8 @@ static bool pfmemalloc_watermark_ok(pg_data_t *pgdat) for (i = 0; i <= ZONE_NORMAL; i++) { zone = &pgdat->node_zones[i]; - if (!populated_zone(zone)) + if (!populated_zone(zone) || + zone_reclaimable_pages(zone) == 0) continue; pfmemalloc_reserve += min_wmark_pages(zone); @@ -3596,7 +3597,7 @@ int zone_reclaim_mode __read_mostly; #define RECLAIM_OFF 0 #define RECLAIM_ZONE (1<<0) /* Run shrink_inactive_list on the zone */ #define RECLAIM_WRITE (1<<1) /* Writeout pages during reclaim */ -#define RECLAIM_SWAP (1<<2) /* Swap pages out during reclaim */ +#define RECLAIM_UNMAP (1<<2) /* Unmap pages during reclaim */ /* * Priority for ZONE_RECLAIM. This determines the fraction of pages @@ -3638,12 +3639,12 @@ static long zone_pagecache_reclaimable(struct zone *zone) long delta = 0; /* - * If RECLAIM_SWAP is set, then all file pages are considered + * If RECLAIM_UNMAP is set, then all file pages are considered * potentially reclaimable. Otherwise, we have to worry about * pages like swapcache and zone_unmapped_file_pages() provides * a better estimate */ - if (zone_reclaim_mode & RECLAIM_SWAP) + if (zone_reclaim_mode & RECLAIM_UNMAP) nr_pagecache_reclaimable = zone_page_state(zone, NR_FILE_PAGES); else nr_pagecache_reclaimable = zone_unmapped_file_pages(zone); @@ -3674,15 +3675,15 @@ static int __zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order) .order = order, .priority = ZONE_RECLAIM_PRIORITY, .may_writepage = !!(zone_reclaim_mode & RECLAIM_WRITE), - .may_unmap = !!(zone_reclaim_mode & RECLAIM_SWAP), + .may_unmap = !!(zone_reclaim_mode & RECLAIM_UNMAP), .may_swap = 1, }; cond_resched(); /* - * We need to be able to allocate from the reserves for RECLAIM_SWAP + * We need to be able to allocate from the reserves for RECLAIM_UNMAP * and we also need to be able to write out pages for RECLAIM_WRITE - * and RECLAIM_SWAP. + * and RECLAIM_UNMAP. */ p->flags |= PF_MEMALLOC | PF_SWAPWRITE; lockdep_set_current_reclaim_state(gfp_mask); |