Patch "mm: numa: avoid waiting on freed migrated pages" has been added to the 4.11-stable tree

This is a note to let you know that I've just added the patch titled

mm: numa: avoid waiting on freed migrated pages

to the 4.11-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary

The filename of the patch is:
mm-numa-avoid-waiting-on-freed-migrated-pages.patch
and it can be found in the queue-4.11 subdirectory.

If you, or anyone else, feels it should not be added to the stable tree, please let <[email protected]> know about it.

From: Mark Rutland <[email protected]>
Date: Fri, 16 Jun 2017 14:02:34 -0700
Subject: mm: numa: avoid waiting on freed migrated pages

From: Mark Rutland <[email protected]>

commit 3c226c637b69104f6b9f1c6ec5b08d7b741b3229 upstream.

In do_huge_pmd_numa_page(), we attempt to handle a migrating thp pmd by waiting until the pmd is unlocked before we return and retry. However, we can race with migrate_misplaced_transhuge_page():

    // do_huge_pmd_numa_page                // migrate_misplaced_transhuge_page()
    // Holds 0 refs on page                 // Holds 2 refs on page

    vmf->ptl = pmd_lock(vma->vm_mm, vmf->pmd);
    /* ... */
    if (pmd_trans_migrating(*vmf->pmd)) {
            page = pmd_page(*vmf->pmd);
            spin_unlock(vmf->ptl);
                                            ptl = pmd_lock(mm, pmd);
                                            if (page_count(page) != 2)) {
                                                    /* roll back */
                                            }
                                            /* ... */
                                            mlock_migrate_page(new_page, page);
                                            /* ... */
                                            spin_unlock(ptl);
                                            put_page(page);
                                            put_page(page); // page freed here
            wait_on_page_locked(page);
            goto out;
    }

This can result in the freed page having its waiters flag set unexpectedly, which trips the PAGE_FLAGS_CHECK_AT_PREP checks in the page alloc/free functions. This has been observed on arm64 KVM guests.

We can avoid this by having do_huge_pmd_numa_page() take a reference on the page before dropping the pmd lock, mirroring what we do in __migration_entry_wait().

When we hit the race, migrate_misplaced_transhuge_page() will see the reference and abort the migration, as it may do today in other cases.

---
 mm/huge_memory.c |    8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

@@ -1412,8 +1412,11 @@ int do_huge_pmd_numa_page(struct vm_faul
 	 */
 	if (unlikely(pmd_trans_migrating(*vmf->pmd))) {
 		page = pmd_page(*vmf->pmd);
+		if (!get_page_unless_zero(page))
+			goto out_unlock;
 		spin_unlock(vmf->ptl);
 		wait_on_page_locked(page);
+		put_page(page);
 		goto out;
 	}
 

@@ -1445,9 +1448,12 @@ int do_huge_pmd_numa_page(struct vm_faul
 
 	/* Migration could have started since the pmd_trans_migrating check */
 	if (!page_locked) {
+		page_nid = -1;
+		if (!get_page_unless_zero(page))
+			goto out_unlock;
 		spin_unlock(vmf->ptl);
 		wait_on_page_locked(page);
-		page_nid = -1;
+		put_page(page);
 		goto out;
 	}
 

Patches currently in stable-queue which might be from [email protected] are

queue-4.11/mm-numa-avoid-waiting-on-freed-migrated-pages.patch