|| ||Minchan Kim <email@example.com> |
|| ||Andrew Morton <firstname.lastname@example.org> |
|| ||[PATCH v4 0/7] f/madivse(DONTNEED) support |
|| ||Mon, 6 Dec 2010 02:29:08 +0900|
|| ||Rik van Riel <email@example.com>,
KOSAKI Motohiro <firstname.lastname@example.org>,
Minchan Kim <email@example.com>|
|| ||Article, Thread
Recently there is a report about working set page eviction due to rsync
workload. application programmers want to use fadvise but it's not easy.
You could see detailed description on [2/7].
- [1/7] is to remove checkpatch's reporting in mm/swap.c
- [2/7] is to move invalidated page which is dirty/writeback on active list
into inactive list's head.
- [3/7] is to move memcg reclaimable page on inactive's tail.
- [4/7] is for moving invalidated page into inactive list's tail when the
page's writeout is completed for reclaim asap.
- [5/7] is to add profing information for evaluation.
- [6/7] is to remove zap_detail NULL dependency to some functions. It is for
- [7/7] is to not calling mark_page_accessed in case of madvise(DONTNEED)
This patches are based on mmotm-12-02
Before applying the series, Please, remove below patches.
Minchan Kim (7):
Fix checkpatch's report in swap.c
deactivate invalidated pages
move memcg reclaimable page into tail of inactive list
Reclaim invalidated page ASAP
add profile information for invalidated page reclaim
Remove zap_details NULL dependency
Prevent activation of page in madvise_dontneed
include/linux/memcontrol.h | 6 ++
include/linux/mm.h | 10 ++++
include/linux/swap.h | 1 +
include/linux/vmstat.h | 4 +-
mm/madvise.c | 13 +++--
mm/memcontrol.c | 27 +++++++++
mm/memory.c | 19 ++++---
mm/mmap.c | 6 ++-
mm/page-writeback.c | 12 ++++-
mm/swap.c | 127 +++++++++++++++++++++++++++++++++++++++++---
mm/truncate.c | 17 +++++--
mm/vmstat.c | 3 +
12 files changed, 216 insertions(+), 29 deletions(-)
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to firstname.lastname@example.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/