This is a note to let you know that I've just added the patch titled
mm: numa: defer TLB flush for THP migration as long as possible
to the 3.12-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-defer-tlb-flush-for-thp-migration-as-long-as-possible.patch
and it can be found in the queue-3.12 subdirectory.
If you, or anyone else, feels it should not be added to the stable tree,
please let <[email protected]> know about it.
>From [email protected] Tue Jan 7 10:51:22 2014
From: Mel Gorman <[email protected]>
Date: Tue, 7 Jan 2014 14:00:48 +0000
Subject: mm: numa: defer TLB flush for THP migration as long as possible
To: [email protected]
Cc: [email protected], [email protected], [email protected], Mel Gorman
<[email protected]>, [email protected]
Message-ID: <[email protected]>
From: Mel Gorman <[email protected]>
commit b0943d61b8fa420180f92f64ef67662b4f6cc493 upstream.
THP migration can fail for a variety of reasons. Avoid flushing the TLB
to deal with THP migration races until the copy is ready to start.
Signed-off-by: Mel Gorman <[email protected]>
Reviewed-by: Rik van Riel <[email protected]>
Cc: Alex Thorlton <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
---
mm/huge_memory.c | 7 -------
mm/migrate.c | 3 +++
2 files changed, 3 insertions(+), 7 deletions(-)
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -1360,13 +1360,6 @@ int do_huge_pmd_numa_page(struct mm_stru
}
/*
- * The page_table_lock above provides a memory barrier
- * with change_protection_range.
- */
- if (mm_tlb_flush_pending(mm))
- flush_tlb_range(vma, haddr, haddr + HPAGE_PMD_SIZE);
-
- /*
* Migrate the THP to the requested node, returns with page unlocked
* and pmd_numa cleared.
*/
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -1705,6 +1705,9 @@ int migrate_misplaced_transhuge_page(str
goto out_fail;
}
+ if (mm_tlb_flush_pending(mm))
+ flush_tlb_range(vma, mmun_start, mmun_end);
+
/* Prepare a page as a migration target */
__set_page_locked(new_page);
SetPageSwapBacked(new_page);
Patches currently in stable-queue which might be from [email protected] are
queue-3.12/mm-numa-clear-numa-hinting-information-on-mprotect.patch
queue-3.12/sched-numa-skip-inaccessible-vmas.patch
queue-3.12/mm-numa-defer-tlb-flush-for-thp-migration-as-long-as-possible.patch
queue-3.12/mm-numa-avoid-unnecessary-disruption-of-numa-hinting-during-migration.patch
queue-3.12/mm-compaction-respect-ignore_skip_hint-in-update_pageblock_skip.patch
queue-3.12/mm-clear-pmd_numa-before-invalidating.patch
queue-3.12/mm-fix-tlb-flush-race-between-migration-and-change_protection_range.patch
queue-3.12/mm-numa-ensure-anon_vma-is-locked-to-prevent-parallel-thp-splits.patch
queue-3.12/mm-numa-do-not-clear-pte-for-pte_numa-update.patch
queue-3.12/mm-hugetlb-check-for-pte-null-pointer-in-__page_check_address.patch
queue-3.12/mm-numa-serialise-parallel-get_user_page-against-thp-migration.patch
queue-3.12/mm-munlock-fix-deadlock-in-__munlock_pagevec.patch
queue-3.12/mm-numa-guarantee-that-tlb_flush_pending-updates-are-visible-before-page-table-updates.patch
queue-3.12/mm-numa-avoid-unnecessary-work-on-the-failure-path.patch
queue-3.12/mm-page_alloc-revert-numa-aspect-of-fair-allocation-policy.patch
queue-3.12/mm-mempolicy-correct-putback-method-for-isolate-pages-if-failed.patch
queue-3.12/mm-numa-do-not-clear-pmd-during-pte-update-scan.patch
queue-3.12/mm-numa-call-mmu-notifiers-on-thp-migration.patch
queue-3.12/mm-munlock-fix-a-bug-where-thp-tail-page-is-encountered.patch
--
To unsubscribe from this list: send the line "unsubscribe stable" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
