1 files changed, 83 insertions, 54 deletions

diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index 626e93db28ba..078832cf3636 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -67,6 +67,7 @@ static unsigned int khugepaged_max_ptes_none __read_mostly = HPAGE_PMD_NR-1;
 
 static int khugepaged(void *none);
 static int khugepaged_slab_init(void);
+static void khugepaged_slab_exit(void);
 
 #define MM_SLOTS_HASH_BITS 10
 static __read_mostly DEFINE_HASHTABLE(mm_slots_hash, MM_SLOTS_HASH_BITS);
@@ -109,9 +110,6 @@ static int set_recommended_min_free_kbytes(void)
 	int nr_zones = 0;
 	unsigned long recommended_min;
 
-	if (!khugepaged_enabled())
-		return 0;
-
 	for_each_populated_zone(zone)
 		nr_zones++;
 
@@ -143,9 +141,8 @@ static int set_recommended_min_free_kbytes(void)
 	setup_per_zone_wmarks();
 	return 0;
 }
-late_initcall(set_recommended_min_free_kbytes);
 
-static int start_khugepaged(void)
+static int start_stop_khugepaged(void)
 {
 	int err = 0;
 	if (khugepaged_enabled()) {
@@ -156,6 +153,7 @@ static int start_khugepaged(void)
 			pr_err("khugepaged: kthread_run(khugepaged) failed\n");
 			err = PTR_ERR(khugepaged_thread);
 			khugepaged_thread = NULL;
+			goto fail;
 		}
 
 		if (!list_empty(&khugepaged_scan.mm_head))
@@ -166,7 +164,7 @@ static int start_khugepaged(void)
 		kthread_stop(khugepaged_thread);
 		khugepaged_thread = NULL;
 	}
-
+fail:
 	return err;
 }
 
@@ -183,7 +181,7 @@ static struct page *get_huge_zero_page(void)
 	struct page *zero_page;
 retry:
 	if (likely(atomic_inc_not_zero(&huge_zero_refcount)))
-		return ACCESS_ONCE(huge_zero_page);
+		return READ_ONCE(huge_zero_page);
 
 	zero_page = alloc_pages((GFP_TRANSHUGE | __GFP_ZERO) & ~__GFP_MOVABLE,
 			HPAGE_PMD_ORDER);
@@ -202,7 +200,7 @@ retry:
 	/* We take additional reference here. It will be put back by shrinker */
 	atomic_set(&huge_zero_refcount, 2);
 	preempt_enable();
-	return ACCESS_ONCE(huge_zero_page);
+	return READ_ONCE(huge_zero_page);
 }
 
 static void put_huge_zero_page(void)
@@ -300,7 +298,7 @@ static ssize_t enabled_store(struct kobject *kobj,
 		int err;
 
 		mutex_lock(&khugepaged_mutex);
-		err = start_khugepaged();
+		err = start_stop_khugepaged();
 		mutex_unlock(&khugepaged_mutex);
 
 		if (err)
@@ -634,27 +632,38 @@ static int __init hugepage_init(void)
 
 	err = hugepage_init_sysfs(&hugepage_kobj);
 	if (err)
-		return err;
+		goto err_sysfs;
 
 	err = khugepaged_slab_init();
 	if (err)
-		goto out;
+		goto err_slab;
 
-	register_shrinker(&huge_zero_page_shrinker);
+	err = register_shrinker(&huge_zero_page_shrinker);
+	if (err)
+		goto err_hzp_shrinker;
 
 	/*
 	 * By default disable transparent hugepages on smaller systems,
 	 * where the extra memory used could hurt more than TLB overhead
 	 * is likely to save.  The admin can still enable it through /sys.
 	 */
-	if (totalram_pages < (512 << (20 - PAGE_SHIFT)))
+	if (totalram_pages < (512 << (20 - PAGE_SHIFT))) {
 		transparent_hugepage_flags = 0;
+		return 0;
+	}
 
-	start_khugepaged();
+	err = start_stop_khugepaged();
+	if (err)
+		goto err_khugepaged;
 
 	return 0;
-out:
+err_khugepaged:
+	unregister_shrinker(&huge_zero_page_shrinker);
+err_hzp_shrinker:
+	khugepaged_slab_exit();
+err_slab:
 	hugepage_exit_sysfs(hugepage_kobj);
+err_sysfs:
 	return err;
 }
 subsys_initcall(hugepage_init);
@@ -708,7 +717,7 @@ static inline pmd_t mk_huge_pmd(struct page *page, pgprot_t prot)
 static int __do_huge_pmd_anonymous_page(struct mm_struct *mm,
 					struct vm_area_struct *vma,
 					unsigned long haddr, pmd_t *pmd,
-					struct page *page)
+					struct page *page, gfp_t gfp)
 {
 	struct mem_cgroup *memcg;
 	pgtable_t pgtable;
@@ -716,7 +725,7 @@ static int __do_huge_pmd_anonymous_page(struct mm_struct *mm,
 
 	VM_BUG_ON_PAGE(!PageCompound(page), page);
 
-	if (mem_cgroup_try_charge(page, mm, GFP_TRANSHUGE, &memcg))
+	if (mem_cgroup_try_charge(page, mm, gfp, &memcg))
 		return VM_FAULT_OOM;
 
 	pgtable = pte_alloc_one(mm, haddr);
@@ -822,7 +831,7 @@ int do_huge_pmd_anonymous_page(struct mm_struct *mm, struct vm_area_struct *vma,
 		count_vm_event(THP_FAULT_FALLBACK);
 		return VM_FAULT_FALLBACK;
 	}
-	if (unlikely(__do_huge_pmd_anonymous_page(mm, vma, haddr, pmd, page))) {
+	if (unlikely(__do_huge_pmd_anonymous_page(mm, vma, haddr, pmd, page, gfp))) {
 		put_page(page);
 		count_vm_event(THP_FAULT_FALLBACK);
 		return VM_FAULT_FALLBACK;
@@ -1080,6 +1089,7 @@ int do_huge_pmd_wp_page(struct mm_struct *mm, struct vm_area_struct *vma,
 	unsigned long haddr;
 	unsigned long mmun_start;	/* For mmu_notifiers */
 	unsigned long mmun_end;		/* For mmu_notifiers */
+	gfp_t huge_gfp;			/* for allocation and charge */
 
 	ptl = pmd_lockptr(mm, pmd);
 	VM_BUG_ON_VMA(!vma->anon_vma, vma);
@@ -1106,10 +1116,8 @@ int do_huge_pmd_wp_page(struct mm_struct *mm, struct vm_area_struct *vma,
 alloc:
 	if (transparent_hugepage_enabled(vma) &&
 	    !transparent_hugepage_debug_cow()) {
-		gfp_t gfp;
-
-		gfp = alloc_hugepage_gfpmask(transparent_hugepage_defrag(vma), 0);
-		new_page = alloc_hugepage_vma(gfp, vma, haddr, HPAGE_PMD_ORDER);
+		huge_gfp = alloc_hugepage_gfpmask(transparent_hugepage_defrag(vma), 0);
+		new_page = alloc_hugepage_vma(huge_gfp, vma, haddr, HPAGE_PMD_ORDER);
 	} else
 		new_page = NULL;
 
@@ -1130,8 +1138,7 @@ alloc:
 		goto out;
 	}
 
-	if (unlikely(mem_cgroup_try_charge(new_page, mm,
-					   GFP_TRANSHUGE, &memcg))) {
+	if (unlikely(mem_cgroup_try_charge(new_page, mm, huge_gfp, &memcg))) {
 		put_page(new_page);
 		if (page) {
 			split_huge_page(page);
@@ -1231,7 +1238,7 @@ struct page *follow_trans_huge_pmd(struct vm_area_struct *vma,
 					  pmd, _pmd,  1))
 			update_mmu_cache_pmd(vma, addr, pmd);
 	}
-	if ((flags & FOLL_MLOCK) && (vma->vm_flags & VM_LOCKED)) {
+	if ((flags & FOLL_POPULATE) && (vma->vm_flags & VM_LOCKED)) {
 		if (page->mapping && trylock_page(page)) {
 			lru_add_drain();
 			if (page->mapping)
@@ -1260,6 +1267,7 @@ int do_huge_pmd_numa_page(struct mm_struct *mm, struct vm_area_struct *vma,
 	int target_nid, last_cpupid = -1;
 	bool page_locked;
 	bool migrated = false;
+	bool was_writable;
 	int flags = 0;
 
 	/* A PROT_NONE fault should not end up here */
@@ -1291,17 +1299,8 @@ int do_huge_pmd_numa_page(struct mm_struct *mm, struct vm_area_struct *vma,
 		flags |= TNF_FAULT_LOCAL;
 	}
 
-	/*
-	 * Avoid grouping on DSO/COW pages in specific and RO pages
-	 * in general, RO pages shouldn't hurt as much anyway since
-	 * they can be in shared cache state.
-	 *
-	 * FIXME! This checks "pmd_dirty()" as an approximation of
-	 * "is this a read-only page", since checking "pmd_write()"
-	 * is even more broken. We haven't actually turned this into
-	 * a writable page, so pmd_write() will always be false.
-	 */
-	if (!pmd_dirty(pmd))
+	/* See similar comment in do_numa_page for explanation */
+	if (!(vma->vm_flags & VM_WRITE))
 		flags |= TNF_NO_GROUP;
 
 	/*
@@ -1358,12 +1357,17 @@ int do_huge_pmd_numa_page(struct mm_struct *mm, struct vm_area_struct *vma,
 	if (migrated) {
 		flags |= TNF_MIGRATED;
 		page_nid = target_nid;
-	}
+	} else
+		flags |= TNF_MIGRATE_FAIL;
 
 	goto out;
 clear_pmdnuma:
 	BUG_ON(!PageLocked(page));
+	was_writable = pmd_write(pmd);
 	pmd = pmd_modify(pmd, vma->vm_page_prot);
+	pmd = pmd_mkyoung(pmd);
+	if (was_writable)
+		pmd = pmd_mkwrite(pmd);
 	set_pmd_at(mm, haddr, pmdp, pmd);
 	update_mmu_cache_pmd(vma, addr, pmdp);
 	unlock_page(page);
@@ -1487,6 +1491,7 @@ int change_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd,
 
 	if (__pmd_trans_huge_lock(pmd, vma, &ptl) == 1) {
 		pmd_t entry;
+		bool preserve_write = prot_numa && pmd_write(*pmd);
 		ret = 1;
 
 		/*
@@ -1502,9 +1507,11 @@ 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 = pmd_modify(entry, newprot);
+			if (preserve_write)
+				entry = pmd_mkwrite(entry);
 			ret = HPAGE_PMD_NR;
 			set_pmd_at(mm, addr, pmd, entry);
-			BUG_ON(pmd_write(entry));
+			BUG_ON(!preserve_write && pmd_write(entry));
 		}
 		spin_unlock(ptl);
 	}
@@ -1976,6 +1983,11 @@ static int __init khugepaged_slab_init(void)
 	return 0;
 }
 
+static void __init khugepaged_slab_exit(void)
+{
+	kmem_cache_destroy(mm_slot_cache);
+}
+
 static inline struct mm_slot *alloc_mm_slot(void)
 {
 	if (!mm_slot_cache)	/* initialization failed */
@@ -2109,7 +2121,7 @@ static void release_pte_pages(pte_t *pte, pte_t *_pte)
 {
 	while (--_pte >= pte) {
 		pte_t pteval = *_pte;
-		if (!pte_none(pteval))
+		if (!pte_none(pteval) && !is_zero_pfn(pte_pfn(pteval)))
 			release_pte_page(pte_page(pteval));
 	}
 }
@@ -2120,13 +2132,13 @@ static int __collapse_huge_page_isolate(struct vm_area_struct *vma,
 {
 	struct page *page;
 	pte_t *_pte;
-	int none = 0;
+	int none_or_zero = 0;
 	bool referenced = false, writable = false;
 	for (_pte = pte; _pte < pte+HPAGE_PMD_NR;
 	     _pte++, address += PAGE_SIZE) {
 		pte_t pteval = *_pte;
-		if (pte_none(pteval)) {
-			if (++none <= khugepaged_max_ptes_none)
+		if (pte_none(pteval) || is_zero_pfn(pte_pfn(pteval))) {
+			if (++none_or_zero <= khugepaged_max_ptes_none)
 				continue;
 			else
 				goto out;
@@ -2207,9 +2219,21 @@ static void __collapse_huge_page_copy(pte_t *pte, struct page *page,
 		pte_t pteval = *_pte;
 		struct page *src_page;
 
-		if (pte_none(pteval)) {
+		if (pte_none(pteval) || is_zero_pfn(pte_pfn(pteval))) {
 			clear_user_highpage(page, address);
 			add_mm_counter(vma->vm_mm, MM_ANONPAGES, 1);
+			if (is_zero_pfn(pte_pfn(pteval))) {
+				/*
+				 * ptl mostly unnecessary.
+				 */
+				spin_lock(ptl);
+				/*
+				 * paravirt calls inside pte_clear here are
+				 * superfluous.
+				 */
+				pte_clear(vma->vm_mm, address, _pte);
+				spin_unlock(ptl);
+			}
 		} else {
 			src_page = pte_page(pteval);
 			copy_user_highpage(page, src_page, address, vma);
@@ -2311,8 +2335,8 @@ static bool khugepaged_prealloc_page(struct page **hpage, bool *wait)
 	return true;
 }
 
-static struct page
-*khugepaged_alloc_page(struct page **hpage, struct mm_struct *mm,
+static struct page *
+khugepaged_alloc_page(struct page **hpage, gfp_t gfp, struct mm_struct *mm,
 		       struct vm_area_struct *vma, unsigned long address,
 		       int node)
 {
@@ -2326,8 +2350,7 @@ static struct page
 	 */
 	up_read(&mm->mmap_sem);
 
-	*hpage = alloc_pages_exact_node(node, alloc_hugepage_gfpmask(
-		khugepaged_defrag(), __GFP_OTHER_NODE), HPAGE_PMD_ORDER);
+	*hpage = alloc_pages_exact_node(node, gfp, HPAGE_PMD_ORDER);
 	if (unlikely(!*hpage)) {
 		count_vm_event(THP_COLLAPSE_ALLOC_FAILED);
 		*hpage = ERR_PTR(-ENOMEM);
@@ -2380,13 +2403,14 @@ static bool khugepaged_prealloc_page(struct page **hpage, bool *wait)
 	return true;
 }
 
-static struct page
-*khugepaged_alloc_page(struct page **hpage, struct mm_struct *mm,
+static struct page *
+khugepaged_alloc_page(struct page **hpage, gfp_t gfp, struct mm_struct *mm,
 		       struct vm_area_struct *vma, unsigned long address,
 		       int node)
 {
 	up_read(&mm->mmap_sem);
 	VM_BUG_ON(!*hpage);
+
 	return  *hpage;
 }
 #endif
@@ -2421,16 +2445,21 @@ static void collapse_huge_page(struct mm_struct *mm,
 	struct mem_cgroup *memcg;
 	unsigned long mmun_start;	/* For mmu_notifiers */
 	unsigned long mmun_end;		/* For mmu_notifiers */
+	gfp_t gfp;
 
 	VM_BUG_ON(address & ~HPAGE_PMD_MASK);
 
+	/* Only allocate from the target node */
+	gfp = alloc_hugepage_gfpmask(khugepaged_defrag(), __GFP_OTHER_NODE) |
+		__GFP_THISNODE;
+
 	/* release the mmap_sem read lock. */
-	new_page = khugepaged_alloc_page(hpage, mm, vma, address, node);
+	new_page = khugepaged_alloc_page(hpage, gfp, mm, vma, address, node);
 	if (!new_page)
 		return;
 
 	if (unlikely(mem_cgroup_try_charge(new_page, mm,
-					   GFP_TRANSHUGE, &memcg)))
+					   gfp, &memcg)))
 		return;
 
 	/*
@@ -2543,7 +2572,7 @@ static int khugepaged_scan_pmd(struct mm_struct *mm,
 {
 	pmd_t *pmd;
 	pte_t *pte, *_pte;
-	int ret = 0, none = 0;
+	int ret = 0, none_or_zero = 0;
 	struct page *page;
 	unsigned long _address;
 	spinlock_t *ptl;
@@ -2561,8 +2590,8 @@ static int khugepaged_scan_pmd(struct mm_struct *mm,
 	for (_address = address, _pte = pte; _pte < pte+HPAGE_PMD_NR;
 	     _pte++, _address += PAGE_SIZE) {
 		pte_t pteval = *_pte;
-		if (pte_none(pteval)) {
-			if (++none <= khugepaged_max_ptes_none)
+		if (pte_none(pteval) || is_zero_pfn(pte_pfn(pteval))) {
+			if (++none_or_zero <= khugepaged_max_ptes_none)
 				continue;
 			else
 				goto out_unmap;