From: Andrew Morton <akpm@linux-foundation.org>
To: Tim Chen <tim.c.chen@linux.intel.com>
Cc: "Huang, Ying" <ying.huang@intel.com>,
dave.hansen@intel.com, ak@linux.intel.com, aaron.lu@intel.com,
linux-mm@kvack.org, linux-kernel@vger.kernel.org,
Hugh Dickins <hughd@google.com>, Shaohua Li <shli@kernel.org>,
Minchan Kim <minchan@kernel.org>, Rik van Riel <riel@redhat.com>,
Andrea Arcangeli <aarcange@redhat.com>,
"Kirill A . Shutemov" <kirill.shutemov@linux.intel.com>,
Vladimir Davydov <vdavydov.dev@gmail.com>,
Johannes Weiner <hannes@cmpxchg.org>,
Michal Hocko <mhocko@kernel.org>,
Hillf Danton <hillf.zj@alibaba-inc.com>,
Christian Borntraeger <borntraeger@de.ibm.com>,
Jonathan Corbet <corbet@lwn.net>
Subject: Re: [PATCH v5 3/9] mm/swap: Split swap cache into 64MB trunks
Date: Wed, 11 Jan 2017 15:09:40 -0800 [thread overview]
Message-ID: <20170111150940.25d951a121a62e1b7eff6f8d@linux-foundation.org> (raw)
In-Reply-To: <735bab895e64c930581ffb0a05b661e01da82bc5.1484082593.git.tim.c.chen@linux.intel.com>
On Wed, 11 Jan 2017 09:55:13 -0800 Tim Chen <tim.c.chen@linux.intel.com> wrote:
> The patch is to improve the scalability of the swap out/in via using
> fine grained locks for the swap cache. In current kernel, one address
> space will be used for each swap device. And in the common
> configuration, the number of the swap device is very small (one is
> typical). This causes the heavy lock contention on the radix tree of
> the address space if multiple tasks swap out/in concurrently. But in
> fact, there is no dependency between pages in the swap cache. So that,
> we can split the one shared address space for each swap device into
> several address spaces to reduce the lock contention. In the patch, the
> shared address space is split into 64MB trunks. 64MB is chosen to
> balance the memory space usage and effect of lock contention reduction.
>
> The size of struct address_space on x86_64 architecture is 408B, so with
> the patch, 6528B more memory will be used for every 1GB swap space on
> x86_64 architecture.
>
> One address space is still shared for the swap entries in the same 64M
> trunks. To avoid lock contention for the first round of swap space
> allocation, the order of the swap clusters in the initial free clusters
> list is changed. The swap space distance between the consecutive swap
> clusters in the free cluster list is at least 64M. After the first
> round of allocation, the swap clusters are expected to be freed
> randomly, so the lock contention should be reduced effectively.
Switching from a single radix-tree to an array of radix-trees to reduce
contention seems a bit hacky. That we can do this and have everything
continue to work tells me that we're simply using an inappropriate data
structure to hold this info.
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2017-01-11 23:09 UTC|newest]
Thread overview: 33+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-01-11 17:55 [PATCH v5 0/9] mm/swap: Regular page swap optimizations Tim Chen
2017-01-11 17:55 ` [PATCH v5 1/9] mm/swap: Fix kernel message in swap_info_get() Tim Chen
2017-01-11 17:55 ` [PATCH v5 2/9] mm/swap: Add cluster lock Tim Chen
2017-01-11 23:00 ` Andrew Morton
2017-01-11 23:07 ` Jonathan Corbet
2017-01-11 23:15 ` Andrew Morton
2017-01-12 1:47 ` Huang, Ying
2017-01-12 1:58 ` Andrew Morton
2017-01-12 2:51 ` Huang, Ying
2017-01-14 4:37 ` [Update][PATCH " Huang, Ying
2017-01-12 1:23 ` [PATCH " Huang, Ying
2017-01-11 17:55 ` [PATCH v5 3/9] mm/swap: Split swap cache into 64MB trunks Tim Chen
2017-01-11 23:09 ` Andrew Morton [this message]
2017-01-11 23:19 ` Andi Kleen
2017-01-12 16:47 ` Tim Chen
2017-01-11 17:55 ` [PATCH v5 4/9] mm/swap: skip read ahead for unreferenced swap slots Tim Chen
2017-01-11 17:55 ` [PATCH v5 5/9] mm/swap: Allocate swap slots in batches Tim Chen
2017-01-11 17:55 ` [PATCH v5 6/9] mm/swap: Free swap slots in batch Tim Chen
2017-01-11 17:55 ` [PATCH v5 7/9] mm/swap: Add cache for swap slots allocation Tim Chen
2017-01-17 2:55 ` [Update][PATCH " Huang, Ying
2017-01-17 10:16 ` Michal Hocko
2017-01-17 17:24 ` Chen, Tim C
2017-01-17 20:03 ` Michal Hocko
2017-01-17 20:31 ` Chen, Tim C
2017-01-17 21:42 ` Tim Chen
2017-01-18 12:45 ` Michal Hocko
2017-01-18 18:03 ` Tim Chen
2017-01-18 18:15 ` Michal Hocko
2017-01-11 17:55 ` [PATCH v5 8/9] mm/swap: Enable swap slots cache usage Tim Chen
2017-01-11 17:55 ` [PATCH v5 9/9] mm/swap: Skip readahead only when swap slot cache is enabled Tim Chen
2017-01-16 12:02 ` [PATCH v5 0/9] mm/swap: Regular page swap optimizations Michal Hocko
2017-01-17 1:06 ` Huang, Ying
2017-01-17 7:49 ` Michal Hocko
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20170111150940.25d951a121a62e1b7eff6f8d@linux-foundation.org \
--to=akpm@linux-foundation.org \
--cc=aarcange@redhat.com \
--cc=aaron.lu@intel.com \
--cc=ak@linux.intel.com \
--cc=borntraeger@de.ibm.com \
--cc=corbet@lwn.net \
--cc=dave.hansen@intel.com \
--cc=hannes@cmpxchg.org \
--cc=hillf.zj@alibaba-inc.com \
--cc=hughd@google.com \
--cc=kirill.shutemov@linux.intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mhocko@kernel.org \
--cc=minchan@kernel.org \
--cc=riel@redhat.com \
--cc=shli@kernel.org \
--cc=tim.c.chen@linux.intel.com \
--cc=vdavydov.dev@gmail.com \
--cc=ying.huang@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).