The patch titled Subject: mm-compaction-acquire-the-zone-lru_lock-as-late-as-possible-fix-fix has been removed from the -mm tree. Its filename was mm-compaction-acquire-the-zone-lru_lock-as-late-as-possible-fix-fix.patch This patch was dropped because it was folded into mm-compaction-acquire-the-zone-lru_lock-as-late-as-possible.patch ------------------------------------------------------ From: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> Subject: mm-compaction-acquire-the-zone-lru_lock-as-late-as-possible-fix-fix Cc: Mel Gorman <mgorman@xxxxxxx> Cc: Minchan Kim <minchan@xxxxxxxxxx> Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> --- mm/compaction.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff -puN mm/compaction.c~mm-compaction-acquire-the-zone-lru_lock-as-late-as-possible-fix-fix mm/compaction.c --- a/mm/compaction.c~mm-compaction-acquire-the-zone-lru_lock-as-late-as-possible-fix-fix +++ a/mm/compaction.c @@ -414,9 +414,10 @@ isolate_migratepages_range(struct zone * * splitting and collapsing (collapsing has already happened * if PageLRU is set) but the lock is not necessarily taken * here and it is wasteful to take it just to check transhuge. - * Check transhuge without lock and skip if it's either a - * transhuge or hugetlbfs page because calling compound_order() - * requires lru_lock to exclude isolation and splitting. + * Check TransHuge without lock and skip the whole pageblock if + * it's either a transhuge or hugetlbfs page, as calling + * compound_order() without preventing THP from splitting the + * page underneath us may return surprising results. */ if (PageTransHuge(page)) { if (!locked) _ Patches currently in -mm which might be from akpm@xxxxxxxxxxxxxxxxxxxx are origin.patch x86-pat-separate-the-pfn-attribute-tracking-for-remap_pfn_range-and-vm_insert_pfn.patch mm-kill-vma-flag-vm_reserved-and-mm-reserved_vm-counter.patch mm-fix-nonuniform-page-status-when-writing-new-file-with-small-buffer.patch mm-vmscan-scale-number-of-pages-reclaimed-by-reclaim-compaction-based-on-failures.patch mm-mmu_notifier-fix-inconsistent-memory-between-secondary-mmu-and-host.patch rbtree-performance-and-correctness-test.patch mm-cma-discard-clean-pages-during-contiguous-allocation-instead-of-migration.patch mm-fix-tracing-in-free_pcppages_bulk.patch cma-count-free-cma-pages.patch memory-hotplug-fix-pages-missed-by-race-rather-than-failing.patch atomic-implement-generic-atomic_dec_if_positive.patch mm-compaction-abort-compaction-loop-if-lock-is-contended-or-run-too-long.patch mm-compaction-acquire-the-zone-lru_lock-as-late-as-possible.patch mm-page_alloc-refactor-out-__alloc_contig_migrate_alloc-checkpatch-fixes.patch mm-fix-invalidate_complete_page2-lock-ordering.patch mm-numa-reclaim-from-all-nodes-within-reclaim-distance-fix.patch mm-move-all-mmu-notifier-invocations-to-be-done-outside-the-pt-lock.patch cma-migrate-mlocked-pages-fix.patch mm-avoid-section-mismatch-warning-for-memblock_type_name-fix.patch mm-document-pagehuge-somewhat.patch mm-memcg-clean-up-mm_match_cgroup-signature-fix.patch memory-hotplug-suppress-trying-to-free-nonexistent-resource-xxxxxxxxxxxxxxxx-yyyyyyyyyyyyyyyy-warning.patch mm-add-and-use-update_mmu_cache_pmd-in-transparent-huge-page-code-fix.patch -- To unsubscribe from this list: send the line "unsubscribe mm-commits" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html