1 files changed, 72 insertions, 36 deletions

diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index d36b2af4d1bf..f3c4f9d22821 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -240,18 +240,18 @@ static ssize_t defrag_store(struct kobject *kobj,
 		clear_bit(TRANSPARENT_HUGEPAGE_DEFRAG_KSWAPD_OR_MADV_FLAG, &transparent_hugepage_flags);
 		clear_bit(TRANSPARENT_HUGEPAGE_DEFRAG_REQ_MADV_FLAG, &transparent_hugepage_flags);
 		set_bit(TRANSPARENT_HUGEPAGE_DEFRAG_DIRECT_FLAG, &transparent_hugepage_flags);
-	} else if (!memcmp("defer", buf,
-		    min(sizeof("defer")-1, count))) {
-		clear_bit(TRANSPARENT_HUGEPAGE_DEFRAG_DIRECT_FLAG, &transparent_hugepage_flags);
-		clear_bit(TRANSPARENT_HUGEPAGE_DEFRAG_KSWAPD_OR_MADV_FLAG, &transparent_hugepage_flags);
-		clear_bit(TRANSPARENT_HUGEPAGE_DEFRAG_REQ_MADV_FLAG, &transparent_hugepage_flags);
-		set_bit(TRANSPARENT_HUGEPAGE_DEFRAG_KSWAPD_FLAG, &transparent_hugepage_flags);
 	} else if (!memcmp("defer+madvise", buf,
 		    min(sizeof("defer+madvise")-1, count))) {
 		clear_bit(TRANSPARENT_HUGEPAGE_DEFRAG_DIRECT_FLAG, &transparent_hugepage_flags);
 		clear_bit(TRANSPARENT_HUGEPAGE_DEFRAG_KSWAPD_FLAG, &transparent_hugepage_flags);
 		clear_bit(TRANSPARENT_HUGEPAGE_DEFRAG_REQ_MADV_FLAG, &transparent_hugepage_flags);
 		set_bit(TRANSPARENT_HUGEPAGE_DEFRAG_KSWAPD_OR_MADV_FLAG, &transparent_hugepage_flags);
+	} else if (!memcmp("defer", buf,
+		    min(sizeof("defer")-1, count))) {
+		clear_bit(TRANSPARENT_HUGEPAGE_DEFRAG_DIRECT_FLAG, &transparent_hugepage_flags);
+		clear_bit(TRANSPARENT_HUGEPAGE_DEFRAG_KSWAPD_OR_MADV_FLAG, &transparent_hugepage_flags);
+		clear_bit(TRANSPARENT_HUGEPAGE_DEFRAG_REQ_MADV_FLAG, &transparent_hugepage_flags);
+		set_bit(TRANSPARENT_HUGEPAGE_DEFRAG_KSWAPD_FLAG, &transparent_hugepage_flags);
 	} else if (!memcmp("madvise", buf,
 			   min(sizeof("madvise")-1, count))) {
 		clear_bit(TRANSPARENT_HUGEPAGE_DEFRAG_DIRECT_FLAG, &transparent_hugepage_flags);
@@ -1568,8 +1568,7 @@ bool madvise_free_huge_pmd(struct mmu_gather *tlb, struct vm_area_struct *vma,
 		deactivate_page(page);
 
 	if (pmd_young(orig_pmd) || pmd_dirty(orig_pmd)) {
-		orig_pmd = pmdp_huge_get_and_clear_full(tlb->mm, addr, pmd,
-			tlb->fullmm);
+		pmdp_invalidate(vma, addr, pmd);
 		orig_pmd = pmd_mkold(orig_pmd);
 		orig_pmd = pmd_mkclean(orig_pmd);
 
@@ -1724,37 +1723,69 @@ int change_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd,
 {
 	struct mm_struct *mm = vma->vm_mm;
 	spinlock_t *ptl;
-	int ret = 0;
+	pmd_t entry;
+	bool preserve_write;
+	int ret;
 
 	ptl = __pmd_trans_huge_lock(pmd, vma);
-	if (ptl) {
-		pmd_t entry;
-		bool preserve_write = prot_numa && pmd_write(*pmd);
-		ret = 1;
+	if (!ptl)
+		return 0;
 
-		/*
-		 * Avoid trapping faults against the zero page. The read-only
-		 * data is likely to be read-cached on the local CPU and
-		 * local/remote hits to the zero page are not interesting.
-		 */
-		if (prot_numa && is_huge_zero_pmd(*pmd)) {
-			spin_unlock(ptl);
-			return ret;
-		}
+	preserve_write = prot_numa && pmd_write(*pmd);
+	ret = 1;
 
-		if (!prot_numa || !pmd_protnone(*pmd)) {
-			entry = pmdp_huge_get_and_clear_notify(mm, addr, pmd);
-			entry = pmd_modify(entry, newprot);
-			if (preserve_write)
-				entry = pmd_mk_savedwrite(entry);
-			ret = HPAGE_PMD_NR;
-			set_pmd_at(mm, addr, pmd, entry);
-			BUG_ON(vma_is_anonymous(vma) && !preserve_write &&
-					pmd_write(entry));
-		}
-		spin_unlock(ptl);
-	}
+	/*
+	 * Avoid trapping faults against the zero page. The read-only
+	 * data is likely to be read-cached on the local CPU and
+	 * local/remote hits to the zero page are not interesting.
+	 */
+	if (prot_numa && is_huge_zero_pmd(*pmd))
+		goto unlock;
+
+	if (prot_numa && pmd_protnone(*pmd))
+		goto unlock;
+
+	/*
+	 * In case prot_numa, we are under down_read(mmap_sem). It's critical
+	 * to not clear pmd intermittently to avoid race with MADV_DONTNEED
+	 * which is also under down_read(mmap_sem):
+	 *
+	 *	CPU0:				CPU1:
+	 *				change_huge_pmd(prot_numa=1)
+	 *				 pmdp_huge_get_and_clear_notify()
+	 * madvise_dontneed()
+	 *  zap_pmd_range()
+	 *   pmd_trans_huge(*pmd) == 0 (without ptl)
+	 *   // skip the pmd
+	 *				 set_pmd_at();
+	 *				 // pmd is re-established
+	 *
+	 * The race makes MADV_DONTNEED miss the huge pmd and don't clear it
+	 * which may break userspace.
+	 *
+	 * pmdp_invalidate() is required to make sure we don't miss
+	 * dirty/young flags set by hardware.
+	 */
+	entry = *pmd;
+	pmdp_invalidate(vma, addr, pmd);
+
+	/*
+	 * Recover dirty/young flags.  It relies on pmdp_invalidate to not
+	 * corrupt them.
+	 */
+	if (pmd_dirty(*pmd))
+		entry = pmd_mkdirty(entry);
+	if (pmd_young(*pmd))
+		entry = pmd_mkyoung(entry);
 
+	entry = pmd_modify(entry, newprot);
+	if (preserve_write)
+		entry = pmd_mk_savedwrite(entry);
+	ret = HPAGE_PMD_NR;
+	set_pmd_at(mm, addr, pmd, entry);
+	BUG_ON(vma_is_anonymous(vma) && !preserve_write && pmd_write(entry));
+unlock:
+	spin_unlock(ptl);
 	return ret;
 }
 
@@ -1828,7 +1859,7 @@ static void __split_huge_pud_locked(struct vm_area_struct *vma, pud_t *pud,
 	VM_BUG_ON_VMA(vma->vm_end < haddr + HPAGE_PUD_SIZE, vma);
 	VM_BUG_ON(!pud_trans_huge(*pud) && !pud_devmap(*pud));
 
-	count_vm_event(THP_SPLIT_PMD);
+	count_vm_event(THP_SPLIT_PUD);
 
 	pudp_huge_clear_flush_notify(vma, haddr, pud);
 }
@@ -2048,6 +2079,7 @@ void split_huge_pmd_address(struct vm_area_struct *vma, unsigned long address,
 		bool freeze, struct page *page)
 {
 	pgd_t *pgd;
+	p4d_t *p4d;
 	pud_t *pud;
 	pmd_t *pmd;
 
@@ -2055,7 +2087,11 @@ void split_huge_pmd_address(struct vm_area_struct *vma, unsigned long address,
 	if (!pgd_present(*pgd))
 		return;
 
-	pud = pud_offset(pgd, address);
+	p4d = p4d_offset(pgd, address);
+	if (!p4d_present(*p4d))
+		return;
+
+	pud = pud_offset(p4d, address);
 	if (!pud_present(*pud))
 		return;