diff options
Diffstat (limited to 'mm/hugetlb.c')
| -rw-r--r-- | mm/hugetlb.c | 195 |
1 files changed, 150 insertions, 45 deletions
diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 924553aa8f78..95dc7b83381f 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -1072,6 +1072,8 @@ static void enqueue_huge_page(struct hstate *h, struct page *page) int nid = page_to_nid(page); lockdep_assert_held(&hugetlb_lock); + VM_BUG_ON_PAGE(page_count(page), page); + list_move(&page->lru, &h->hugepage_freelists[nid]); h->free_huge_pages++; h->free_huge_pages_node[nid]++; @@ -1143,7 +1145,7 @@ static struct page *dequeue_huge_page_vma(struct hstate *h, unsigned long address, int avoid_reserve, long chg) { - struct page *page; + struct page *page = NULL; struct mempolicy *mpol; gfp_t gfp_mask; nodemask_t *nodemask; @@ -1164,7 +1166,17 @@ static struct page *dequeue_huge_page_vma(struct hstate *h, gfp_mask = htlb_alloc_mask(h); nid = huge_node(vma, address, gfp_mask, &mpol, &nodemask); - page = dequeue_huge_page_nodemask(h, gfp_mask, nid, nodemask); + + if (mpol_is_preferred_many(mpol)) { + page = dequeue_huge_page_nodemask(h, gfp_mask, nid, nodemask); + + /* Fallback to all nodes if page==NULL */ + nodemask = NULL; + } + + if (!page) + page = dequeue_huge_page_nodemask(h, gfp_mask, nid, nodemask); + if (page && !avoid_reserve && vma_has_reserves(vma, chg)) { SetHPageRestoreReserve(page); h->resv_huge_pages--; @@ -1368,8 +1380,28 @@ static void remove_hugetlb_page(struct hstate *h, struct page *page, h->surplus_huge_pages_node[nid]--; } + /* + * Very subtle + * + * For non-gigantic pages set the destructor to the normal compound + * page dtor. This is needed in case someone takes an additional + * temporary ref to the page, and freeing is delayed until they drop + * their reference. + * + * For gigantic pages set the destructor to the null dtor. This + * destructor will never be called. Before freeing the gigantic + * page destroy_compound_gigantic_page will turn the compound page + * into a simple group of pages. After this the destructor does not + * apply. + * + * This handles the case where more than one ref is held when and + * after update_and_free_page is called. + */ set_page_refcounted(page); - set_compound_page_dtor(page, NULL_COMPOUND_DTOR); + if (hstate_is_gigantic(h)) + set_compound_page_dtor(page, NULL_COMPOUND_DTOR); + else + set_compound_page_dtor(page, COMPOUND_PAGE_DTOR); h->nr_huge_pages--; h->nr_huge_pages_node[nid]--; @@ -1399,11 +1431,20 @@ static void add_hugetlb_page(struct hstate *h, struct page *page, SetHPageVmemmapOptimized(page); /* - * This page is now managed by the hugetlb allocator and has - * no users -- drop the last reference. + * This page is about to be managed by the hugetlb allocator and + * should have no users. Drop our reference, and check for others + * just in case. */ zeroed = put_page_testzero(page); - VM_BUG_ON_PAGE(!zeroed, page); + if (!zeroed) + /* + * It is VERY unlikely soneone else has taken a ref on + * the page. In this case, we simply return as the + * hugetlb destructor (free_huge_page) will be called + * when this other ref is dropped. + */ + return; + arch_clear_hugepage_flags(page); enqueue_huge_page(h, page); } @@ -1657,16 +1698,14 @@ static bool prep_compound_gigantic_page(struct page *page, unsigned int order) * cache adding could take a ref on a 'to be' tail page. * We need to respect any increased ref count, and only set * the ref count to zero if count is currently 1. If count - * is not 1, we call synchronize_rcu in the hope that a rcu - * grace period will cause ref count to drop and then retry. - * If count is still inflated on retry we return an error and - * must discard the pages. + * is not 1, we return an error. An error return indicates + * the set of pages can not be converted to a gigantic page. + * The caller who allocated the pages should then discard the + * pages using the appropriate free interface. */ if (!page_ref_freeze(p, 1)) { - pr_info("HugeTLB unexpected inflated ref count on freshly allocated page\n"); - synchronize_rcu(); - if (!page_ref_freeze(p, 1)) - goto out_error; + pr_warn("HugeTLB page can not be used due to unexpected inflated ref count\n"); + goto out_error; } set_page_count(p, 0); set_compound_head(p, page); @@ -1830,7 +1869,6 @@ retry: retry = true; goto retry; } - pr_warn("HugeTLB page can not be used due to unexpected inflated ref count\n"); return NULL; } } @@ -2020,9 +2058,10 @@ int dissolve_free_huge_pages(unsigned long start_pfn, unsigned long end_pfn) * Allocates a fresh surplus page from the page allocator. */ static struct page *alloc_surplus_huge_page(struct hstate *h, gfp_t gfp_mask, - int nid, nodemask_t *nmask) + int nid, nodemask_t *nmask, bool zero_ref) { struct page *page = NULL; + bool retry = false; if (hstate_is_gigantic(h)) return NULL; @@ -2032,6 +2071,7 @@ static struct page *alloc_surplus_huge_page(struct hstate *h, gfp_t gfp_mask, goto out_unlock; spin_unlock_irq(&hugetlb_lock); +retry: page = alloc_fresh_huge_page(h, gfp_mask, nid, nmask, NULL); if (!page) return NULL; @@ -2049,11 +2089,35 @@ static struct page *alloc_surplus_huge_page(struct hstate *h, gfp_t gfp_mask, spin_unlock_irq(&hugetlb_lock); put_page(page); return NULL; - } else { - h->surplus_huge_pages++; - h->surplus_huge_pages_node[page_to_nid(page)]++; } + if (zero_ref) { + /* + * Caller requires a page with zero ref count. + * We will drop ref count here. If someone else is holding + * a ref, the page will be freed when they drop it. Abuse + * temporary page flag to accomplish this. + */ + SetHPageTemporary(page); + if (!put_page_testzero(page)) { + /* + * Unexpected inflated ref count on freshly allocated + * huge. Retry once. + */ + pr_info("HugeTLB unexpected inflated ref count on freshly allocated page\n"); + spin_unlock_irq(&hugetlb_lock); + if (retry) + return NULL; + + retry = true; + goto retry; + } + ClearHPageTemporary(page); + } + + h->surplus_huge_pages++; + h->surplus_huge_pages_node[page_to_nid(page)]++; + out_unlock: spin_unlock_irq(&hugetlb_lock); @@ -2088,16 +2152,26 @@ static struct page *alloc_buddy_huge_page_with_mpol(struct hstate *h, struct vm_area_struct *vma, unsigned long addr) { - struct page *page; + struct page *page = NULL; struct mempolicy *mpol; gfp_t gfp_mask = htlb_alloc_mask(h); int nid; nodemask_t *nodemask; nid = huge_node(vma, addr, gfp_mask, &mpol, &nodemask); - page = alloc_surplus_huge_page(h, gfp_mask, nid, nodemask); - mpol_cond_put(mpol); + if (mpol_is_preferred_many(mpol)) { + gfp_t gfp = gfp_mask | __GFP_NOWARN; + gfp &= ~(__GFP_DIRECT_RECLAIM | __GFP_NOFAIL); + page = alloc_surplus_huge_page(h, gfp, nid, nodemask, false); + + /* Fallback to all nodes if page==NULL */ + nodemask = NULL; + } + + if (!page) + page = alloc_surplus_huge_page(h, gfp_mask, nid, nodemask, false); + mpol_cond_put(mpol); return page; } @@ -2167,7 +2241,7 @@ retry: spin_unlock_irq(&hugetlb_lock); for (i = 0; i < needed; i++) { page = alloc_surplus_huge_page(h, htlb_alloc_mask(h), - NUMA_NO_NODE, NULL); + NUMA_NO_NODE, NULL, true); if (!page) { alloc_ok = false; break; @@ -2208,24 +2282,20 @@ retry: /* Free the needed pages to the hugetlb pool */ list_for_each_entry_safe(page, tmp, &surplus_list, lru) { - int zeroed; - if ((--needed) < 0) break; - /* - * This page is now managed by the hugetlb allocator and has - * no users -- drop the buddy allocator's reference. - */ - zeroed = put_page_testzero(page); - VM_BUG_ON_PAGE(!zeroed, page); + /* Add the page to the hugetlb allocator */ enqueue_huge_page(h, page); } free: spin_unlock_irq(&hugetlb_lock); - /* Free unnecessary surplus pages to the buddy allocator */ + /* + * Free unnecessary surplus pages to the buddy allocator. + * Pages have no ref count, call free_huge_page directly. + */ list_for_each_entry_safe(page, tmp, &surplus_list, lru) - put_page(page); + free_huge_page(page); spin_lock_irq(&hugetlb_lock); return ret; @@ -2476,7 +2546,7 @@ void restore_reserve_on_error(struct hstate *h, struct vm_area_struct *vma, if (!rc) { /* * This indicates there is an entry in the reserve map - * added by alloc_huge_page. We know it was added + * not added by alloc_huge_page. We know it was added * before the alloc_huge_page call, otherwise * HPageRestoreReserve would be set on the page. * Remove the entry so that a subsequent allocation @@ -2534,6 +2604,7 @@ static int alloc_and_dissolve_huge_page(struct hstate *h, struct page *old_page, { gfp_t gfp_mask = htlb_alloc_mask(h) | __GFP_THISNODE; int nid = page_to_nid(old_page); + bool alloc_retry = false; struct page *new_page; int ret = 0; @@ -2544,9 +2615,30 @@ static int alloc_and_dissolve_huge_page(struct hstate *h, struct page *old_page, * the pool. This simplifies and let us do most of the processing * under the lock. */ +alloc_retry: new_page = alloc_buddy_huge_page(h, gfp_mask, nid, NULL, NULL); if (!new_page) return -ENOMEM; + /* + * If all goes well, this page will be directly added to the free + * list in the pool. For this the ref count needs to be zero. + * Attempt to drop now, and retry once if needed. It is VERY + * unlikely there is another ref on the page. + * + * If someone else has a reference to the page, it will be freed + * when they drop their ref. Abuse temporary page flag to accomplish + * this. Retry once if there is an inflated ref count. + */ + SetHPageTemporary(new_page); + if (!put_page_testzero(new_page)) { + if (alloc_retry) + return -EBUSY; + + alloc_retry = true; + goto alloc_retry; + } + ClearHPageTemporary(new_page); + __prep_new_huge_page(h, new_page); retry: @@ -2586,11 +2678,10 @@ retry: remove_hugetlb_page(h, old_page, false); /* - * Reference count trick is needed because allocator gives us - * referenced page but the pool requires pages with 0 refcount. + * Ref count on new page is already zero as it was dropped + * earlier. It can be directly added to the pool free list. */ __prep_account_new_huge_page(h, nid); - page_ref_dec(new_page); enqueue_huge_page(h, new_page); /* @@ -2604,6 +2695,8 @@ retry: free_new: spin_unlock_irq(&hugetlb_lock); + /* Page has a zero ref count, but needs a ref to be freed */ + set_page_refcounted(new_page); update_and_free_page(h, new_page, false); return ret; @@ -2828,8 +2921,8 @@ static void __init gather_bootmem_prealloc(void) prep_new_huge_page(h, page, page_to_nid(page)); put_page(page); /* add to the hugepage allocator */ } else { + /* VERY unlikely inflated ref count on a tail page */ free_gigantic_page(page, huge_page_order(h)); - pr_warn("HugeTLB page can not be used due to unexpected inflated ref count\n"); } /* @@ -4033,8 +4126,10 @@ static void hugetlb_vm_op_open(struct vm_area_struct *vma) * after this open call completes. It is therefore safe to take a * new reference here without additional locking. */ - if (resv && is_vma_resv_set(vma, HPAGE_RESV_OWNER)) + if (resv && is_vma_resv_set(vma, HPAGE_RESV_OWNER)) { + resv_map_dup_hugetlb_cgroup_uncharge_info(resv); kref_get(&resv->refs); + } } static void hugetlb_vm_op_close(struct vm_area_struct *vma) @@ -4660,7 +4755,9 @@ retry_avoidcopy: spin_unlock(ptl); mmu_notifier_invalidate_range_end(&range); out_release_all: - restore_reserve_on_error(h, vma, haddr, new_page); + /* No restore in case of successful pagetable update (Break COW) */ + if (new_page != old_page) + restore_reserve_on_error(h, vma, haddr, new_page); put_page(new_page); out_release_old: put_page(old_page); @@ -4776,7 +4873,7 @@ static vm_fault_t hugetlb_no_page(struct mm_struct *mm, pte_t new_pte; spinlock_t *ptl; unsigned long haddr = address & huge_page_mask(h); - bool new_page = false; + bool new_page, new_pagecache_page = false; /* * Currently, we are forced to kill the process in the event the @@ -4799,6 +4896,7 @@ static vm_fault_t hugetlb_no_page(struct mm_struct *mm, goto out; retry: + new_page = false; page = find_lock_page(mapping, idx); if (!page) { /* Check for page in userfault range */ @@ -4842,6 +4940,7 @@ retry: goto retry; goto out; } + new_pagecache_page = true; } else { lock_page(page); if (unlikely(anon_vma_prepare(vma))) { @@ -4926,7 +5025,9 @@ backout: spin_unlock(ptl); backout_unlocked: unlock_page(page); - restore_reserve_on_error(h, vma, haddr, page); + /* restore reserve for newly allocated pages not in page cache */ + if (new_page && !new_pagecache_page) + restore_reserve_on_error(h, vma, haddr, page); put_page(page); goto out; } @@ -5135,6 +5236,7 @@ int hugetlb_mcopy_atomic_pte(struct mm_struct *dst_mm, int ret = -ENOMEM; struct page *page; int writable; + bool new_pagecache_page = false; if (is_continue) { ret = -EFAULT; @@ -5228,6 +5330,7 @@ int hugetlb_mcopy_atomic_pte(struct mm_struct *dst_mm, ret = huge_add_to_page_cache(page, mapping, idx); if (ret) goto out_release_nounlock; + new_pagecache_page = true; } ptl = huge_pte_lockptr(h, dst_mm, dst_pte); @@ -5291,7 +5394,8 @@ out_release_unlock: if (vm_shared || is_continue) unlock_page(page); out_release_nounlock: - restore_reserve_on_error(h, dst_vma, dst_addr, page); + if (!new_pagecache_page) + restore_reserve_on_error(h, dst_vma, dst_addr, page); put_page(page); goto out; } @@ -5440,8 +5544,9 @@ long follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma, continue; } - refs = min3(pages_per_huge_page(h) - pfn_offset, - (vma->vm_end - vaddr) >> PAGE_SHIFT, remainder); + /* vaddr may not be aligned to PAGE_SIZE */ + refs = min3(pages_per_huge_page(h) - pfn_offset, remainder, + (vma->vm_end - ALIGN_DOWN(vaddr, PAGE_SIZE)) >> PAGE_SHIFT); if (pages || vmas) record_subpages_vmas(mem_map_offset(page, pfn_offset), |