[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-Id: <1518972475-11340-1-git-send-email-robert.m.harris@oracle.com>
Date: Sun, 18 Feb 2018 16:47:54 +0000
From: robert.m.harris@...cle.com
To: linux-mm@...ck.org, linux-kernel@...r.kernel.org,
linux-doc@...r.kernel.org
Cc: Jonathan Corbet <corbet@....net>,
Andrew Morton <akpm@...ux-foundation.org>,
Michal Hocko <mhocko@...e.com>,
Vlastimil Babka <vbabka@...e.cz>,
"Kirill A. Shutemov" <kirill.shutemov@...ux.intel.com>,
Johannes Weiner <hannes@...xchg.org>,
Kemi Wang <kemi.wang@...el.com>,
David Rientjes <rientjes@...gle.com>,
Yafang Shao <laoar.shao@...il.com>,
Kangmin Park <l4stpr0gr4m@...il.com>,
Mel Gorman <mgorman@...e.de>,
Yisheng Xie <xieyisheng1@...wei.com>,
Davidlohr Bueso <dave@...olabs.net>,
Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
Huang Ying <ying.huang@...el.com>,
Vinayak Menon <vinmenon@...eaurora.org>,
"Robert M. Harris" <robert.m.harris@...cle.com>
Subject: [PATCH 0/1] mm, compaction: correct the bounds of __fragmentation_index()
From: "Robert M. Harris" <robert.m.harris@...cle.com>
__fragmentation_index() calculates a value used to determine whether
compaction should be favoured over page reclaim in the event of
allocation failure. The function purports to return a value between 0
and 1000, representing units of 1/1000. Barring the case of a
pathological shortfall of memory, the lower bound is instead 500. This
is significant because it is the default value of
sysctl_extfrag_threshold, i.e. the value below which compaction should
be avoided in favour of page reclaim for costly pages.
Here's an illustration using a zone that I fragmented with selective
calls to __alloc_pages() and __free_pages --- the fragmentation for
order-1 could not be minimised further yet is reported as 0.5:
# head -1 /proc/buddyinfo
Node 0, zone DMA 1983 0 0 0 0 0 0 0 0 0 0
# head -1 /sys/kernel/debug/extfrag/extfrag_index
Node 0, zone DMA -1.000 0.500 0.750 0.875 0.937 0.969 0.984 0.992 0.996 0.998 0.999
#
With extreme memory shortage the reported fragmentation index does go
lower. In fact, it can go below zero:
# head -1 /proc/buddyinfo
Node 0, zone DMA 1 0 0 0 0 0 0 0 0 0 0
# head -1 /sys/kernel/debug/extfrag/extfrag_index
Node 0, zone DMA -1.000 0.-500 0.-250 0.-125 0.-62 0.-31 0.-15 0.-07 0.-03 0.-01 0.000
#
This patch implements and documents a modified version of the original
expression that returns a value in the range 0 <= index < 1000. It
amends the default value of sysctl_extfrag_threshold to preserve the
existing behaviour. With this patch in place, the same two tests yield
# head -1 /proc/buddyinfo
Node 0, zone DMA 1983 0 0 0 0 0 0 0 0 0 0
# head -1 /sys/kernel/debug/extfrag/extfrag_index
Node 0, zone DMA -1.000 0.000 0.500 0.750 0.875 0.937 0.969 0.984 0.992 0.996 0.998
#
and
# head -1 /proc/buddyinfo
Node 0, zone DMA 1 0 0 0 0 0 0 0 0 0 0
# head -1 /sys/kernel/debug/extfrag/extfrag_index
Node 0, zone DMA -1.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
#
Robert M. Harris (1):
mm, compaction: correct the bounds of __fragmentation_index()
Documentation/sysctl/vm.txt | 2 +-
mm/compaction.c | 2 +-
mm/vmstat.c | 47 +++++++++++++++++++++++++++++++++++----------
3 files changed, 39 insertions(+), 12 deletions(-)
--
1.8.3.1
Powered by blists - more mailing lists