lkml.org 
[lkml]   [2017]   [Feb]   [3]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
Subject[PATCH V2 0/7] mm: fix some MADV_FREE issues
Date
Hi,

We are trying to use MADV_FREE in jemalloc. Several issues are found. Without
solving the issues, jemalloc can't use the MADV_FREE feature.
- Doesn't support system without swap enabled. Because if swap is off, we can't
or can't efficiently age anonymous pages. And since MADV_FREE pages are mixed
with other anonymous pages, we can't reclaim MADV_FREE pages. In current
implementation, MADV_FREE will fallback to MADV_DONTNEED without swap enabled.
But in our environment, a lot of machines don't enable swap. This will prevent
our setup using MADV_FREE.
- Increases memory pressure. page reclaim bias file pages reclaim against
anonymous pages. This doesn't make sense for MADV_FREE pages, because those
pages could be freed easily and refilled with very slight penality. Even page
reclaim doesn't bias file pages, there is still an issue, because MADV_FREE
pages and other anonymous pages are mixed together. To reclaim a MADV_FREE
page, we probably must scan a lot of other anonymous pages, which is
inefficient. In our test, we usually see oom with MADV_FREE enabled and nothing
without it.
- RSS accounting. MADV_FREE pages are accounted as normal anon pages and
reclaimed lazily, so application's RSS becomes bigger. This confuses our
workloads. We have monitoring daemon running and if it finds applications' RSS
becomes abnormal, the daemon will kill the applications even kernel can reclaim
the memory easily. Currently we don't export separate RSS accounting for
MADV_FREE pages. This will prevent our setup using MADV_FREE too.

To address the first the two issues, we can either put MADV_FREE pages into a
separate LRU list (Minchan's previous patches and V1 patches), or put them into
LRU_INACTIVE_FILE list (suggested by Johannes). The patchset use the second
idea. The reason is LRU_INACTIVE_FILE list is tiny nowadays and should be full
of used once file pages. So we can still efficiently reclaim MADV_FREE pages
there without interference with other anon and active file pages. Putting the
pages into inactive file list also has an advantage which allows page reclaim
to prioritize MADV_FREE pages and used once file pages. MADV_FREE pages are put
into the lru list and clear SwapBacked flag, so PageAnon(page) &&
!PageSwapBacked(page) will indicate a MADV_FREE pages. These pages will
directly freed without pageout if they are clean, otherwise normal swap will
reclaim them.

For the third issue, we add a separate RSS count for MADV_FREE pages. The count
will be increased in madvise syscall and decreased in page reclaim (eg, unmap).
There is one limitation, the accounting doesn't work well for shared pages.
Please check the last patch. This probably isn't a big issue, because userspace
will write the pages before reusing them, which will break the page sharing
between two processes. And if two processes share a page, the page can't really
be lazyfreed.

Thanks,
Shaohua

V1->V2:
- Put MADV_FREE pages into LRU_INACTIVE_FILE list instead of adding a new lru
list, suggested by Johannes
- Add RSS support

Minchan previous patches:
http://marc.info/?l=linux-mm&m=144800657002763&w=2
----------------------
Shaohua Li (7):
mm: don't assume anonymous pages have SwapBacked flag
mm: move MADV_FREE pages into LRU_INACTIVE_FILE list
mm: reclaim MADV_FREE pages
mm: enable MADV_FREE for swapless system
mm: add vmstat account for MADV_FREE pages
proc: show MADV_FREE pages info in smaps
mm: add a separate RSS for MADV_FREE pages

drivers/base/node.c | 2 ++
fs/proc/array.c | 9 +++++---
fs/proc/internal.h | 3 ++-
fs/proc/meminfo.c | 1 +
fs/proc/task_mmu.c | 17 +++++++++++---
fs/proc/task_nommu.c | 4 +++-
include/linux/mm_inline.h | 36 +++++++++++++++++++++++++++---
include/linux/mm_types.h | 1 +
include/linux/mmzone.h | 2 ++
include/linux/page-flags.h | 6 +++++
include/linux/swap.h | 2 +-
include/linux/vm_event_item.h | 2 +-
mm/gup.c | 2 ++
mm/huge_memory.c | 14 ++++++++----
mm/khugepaged.c | 10 ++++-----
mm/madvise.c | 16 ++++++-------
mm/memory.c | 13 +++++++++--
mm/migrate.c | 5 ++++-
mm/oom_kill.c | 10 +++++----
mm/page_alloc.c | 7 ++++--
mm/rmap.c | 10 ++++++++-
mm/swap.c | 50 +++++++++++++++++++++++------------------
mm/vmscan.c | 52 +++++++++++++++++++++++++++++++------------
mm/vmstat.c | 3 +++
24 files changed, 200 insertions(+), 77 deletions(-)

--
2.9.3

\
 
 \ /
  Last update: 2017-02-04 00:36    [W:0.349 / U:0.148 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site