diff options
| author | Vlastimil Babka <vbabka@suse.cz> | 2013-09-11 14:22:35 -0700 |
|---|---|---|
| committer | Linus Torvalds <torvalds@linux-foundation.org> | 2013-09-11 15:58:01 -0700 |
| commit | 7a8010cd36273ff5f6fea5201ef9232f30cebbd9 (patch) | |
| tree | 3805f3d9a8a1f1c1c555ef31bc1bdb51fb51e33e /include | |
| parent | 5b40998ae35cf64561868370e6c9f3d3e94b6bf7 (diff) | |
mm: munlock: manual pte walk in fast path instead of follow_page_mask()
Currently munlock_vma_pages_range() calls follow_page_mask() to obtain
each individual struct page. This entails repeated full page table
translations and page table lock taken for each page separately.
This patch avoids the costly follow_page_mask() where possible, by
iterating over ptes within single pmd under single page table lock. The
first pte is obtained by get_locked_pte() for non-THP page acquired by the
initial follow_page_mask(). The rest of the on-stack pagevec for munlock
is filled up using pte_walk as long as pte_present() and vm_normal_page()
are sufficient to obtain the struct page.
After this patch, a 14% speedup was measured for munlocking a 56GB large
memory area with THP disabled.
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Jörn Engel <joern@logfs.org>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Michel Lespinasse <walken@google.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@suse.cz>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Diffstat (limited to 'include')
| -rw-r--r-- | include/linux/mm.h | 12 |
1 files changed, 6 insertions, 6 deletions
diff --git a/include/linux/mm.h b/include/linux/mm.h index dce24569f8f..03f84b8d735 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -643,12 +643,12 @@ static inline enum zone_type page_zonenum(const struct page *page) #endif /* - * The identification function is only used by the buddy allocator for - * determining if two pages could be buddies. We are not really - * identifying a zone since we could be using a the section number - * id if we have not node id available in page flags. - * We guarantee only that it will return the same value for two - * combinable pages in a zone. + * The identification function is mainly used by the buddy allocator for + * determining if two pages could be buddies. We are not really identifying + * the zone since we could be using the section number id if we do not have + * node id available in page flags. + * We only guarantee that it will return the same value for two combinable + * pages in a zone. */ static inline int page_zone_id(struct page *page) { |