mm: numa: serialise parallel get_user_page against THP migration

Base pages are unmapped and flushed from cache and TLB during normal
page migration and replaced with a migration entry that causes any
parallel NUMA hinting fault or gup to block until migration completes.

THP does not unmap pages due to a lack of support for migration entries
at a PMD level.  This allows races with get_user_pages and
get_user_pages_fast which commit 3f926ab945b6 ("mm: Close races between
THP migration and PMD numa clearing") made worse by introducing a
pmd_clear_flush().

This patch forces get_user_page (fast and normal) on a pmd_numa page to
go through the slow get_user_page path where it will serialise against
THP migration and properly account for the NUMA hinting fault.  On the
migration side the page table lock is taken for each PTE update.

Signed-off-by: Mel Gorman <mgorman@suse.de>
Reviewed-by: Rik van Riel <riel@redhat.com>
Cc: Alex Thorlton <athorlton@sgi.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

1 files changed, 31 insertions, 7 deletions

diff --git a/mm/migrate.c b/mm/migrate.c
index bb94004..2cabbd5 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -1722,6 +1722,7 @@ int migrate_misplaced_transhuge_page(struct mm_struct *mm,
 	struct page *new_page = NULL;
 	struct mem_cgroup *memcg = NULL;
 	int page_lru = page_is_file_cache(page);
+	pmd_t orig_entry;
 
 	/*
 	 * Rate-limit the amount of data that is being migrated to a node.
@@ -1756,7 +1757,8 @@ int migrate_misplaced_transhuge_page(struct mm_struct *mm,
 
 	/* Recheck the target PMD */
 	ptl = pmd_lock(mm, pmd);
-	if (unlikely(!pmd_same(*pmd, entry))) {
+	if (unlikely(!pmd_same(*pmd, entry) || page_count(page) != 2)) {
+fail_putback:
 		spin_unlock(ptl);
 
 		/* Reverse changes made by migrate_page_copy() */
@@ -1786,16 +1788,34 @@ int migrate_misplaced_transhuge_page(struct mm_struct *mm,
 	 */
 	mem_cgroup_prepare_migration(page, new_page, &memcg);
 
+	orig_entry = *pmd;
 	entry = mk_pmd(new_page, vma->vm_page_prot);
-	entry = pmd_mknonnuma(entry);
-	entry = maybe_pmd_mkwrite(pmd_mkdirty(entry), vma);
 	entry = pmd_mkhuge(entry);
+	entry = maybe_pmd_mkwrite(pmd_mkdirty(entry), vma);
 
+	/*
+	 * Clear the old entry under pagetable lock and establish the new PTE.
+	 * Any parallel GUP will either observe the old page blocking on the
+	 * page lock, block on the page table lock or observe the new page.
+	 * The SetPageUptodate on the new page and page_add_new_anon_rmap
+	 * guarantee the copy is visible before the pagetable update.
+	 */
+	flush_cache_range(vma, haddr, haddr + HPAGE_PMD_SIZE);
+	page_add_new_anon_rmap(new_page, vma, haddr);
 	pmdp_clear_flush(vma, haddr, pmd);
 	set_pmd_at(mm, haddr, pmd, entry);
-	page_add_new_anon_rmap(new_page, vma, haddr);
 	update_mmu_cache_pmd(vma, address, &entry);
+
+	if (page_count(page) != 2) {
+		set_pmd_at(mm, haddr, pmd, orig_entry);
+		flush_tlb_range(vma, haddr, haddr + HPAGE_PMD_SIZE);
+		update_mmu_cache_pmd(vma, address, &entry);
+		page_remove_rmap(new_page);
+		goto fail_putback;
+	}
+
 	page_remove_rmap(page);
+
 	/*
 	 * Finish the charge transaction under the page table lock to
 	 * prevent split_huge_page() from dividing up the charge
@@ -1820,9 +1840,13 @@ int migrate_misplaced_transhuge_page(struct mm_struct *mm,
 out_fail:
 	count_vm_events(PGMIGRATE_FAIL, HPAGE_PMD_NR);
 out_dropref:
-	entry = pmd_mknonnuma(entry);
-	set_pmd_at(mm, haddr, pmd, entry);
-	update_mmu_cache_pmd(vma, address, &entry);
+	ptl = pmd_lock(mm, pmd);
+	if (pmd_same(*pmd, entry)) {
+		entry = pmd_mknonnuma(entry);
+		set_pmd_at(mm, haddr, pmd, entry);
+		update_mmu_cache_pmd(vma, address, &entry);
+	}
+	spin_unlock(ptl);
 
 	unlock_page(page);
 	put_page(page);