From: Mel Gorman <mgorman@techsingularity.net>
To: Aaron Lu <aaron.lu@intel.com>
Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org,
Andrew Morton <akpm@linux-foundation.org>,
Huang Ying <ying.huang@intel.com>,
Dave Hansen <dave.hansen@intel.com>,
Kemi Wang <kemi.wang@intel.com>,
Tim Chen <tim.c.chen@linux.intel.com>,
Andi Kleen <ak@linux.intel.com>, Michal Hocko <mhocko@suse.com>,
Vlastimil Babka <vbabka@suse.cz>
Subject: Re: [PATCH 2/2] free_pcppages_bulk: prefetch buddy while not holding lock
Date: Wed, 24 Jan 2018 16:43:45 +0000 [thread overview]
Message-ID: <20180124164344.lca63gjn7mefuiac@techsingularity.net> (raw)
In-Reply-To: <20180124023050.20097-2-aaron.lu@intel.com>
On Wed, Jan 24, 2018 at 10:30:50AM +0800, Aaron Lu wrote:
> When a page is freed back to the global pool, its buddy will be checked
> to see if it's possible to do a merge. This requires accessing buddy's
> page structure and that access could take a long time if it's cache cold.
>
> This patch adds a prefetch to the to-be-freed page's buddy outside of
> zone->lock in hope of accessing buddy's page structure later under
> zone->lock will be faster.
>
> Test with will-it-scale/page_fault1 full load:
>
> kernel Broadwell(2S) Skylake(2S) Broadwell(4S) Skylake(4S)
> v4.15-rc4 9037332 8000124 13642741 15728686
> patch1/2 9608786 +6.3% 8368915 +4.6% 14042169 +2.9% 17433559 +10.8%
> this patch 10462292 +8.9% 8602889 +2.8% 14802073 +5.4% 17624575 +1.1%
>
> Note: this patch's performance improvement percent is against patch1/2.
>
I'm less convinced by this for a microbenchmark. Prefetch has not been a
universal win in the past and we cannot be sure that it's a good idea on
all architectures or doesn't have other side-effects such as consuming
memory bandwidth for data we don't need or evicting cache hot data for
buddy information that is not used. Furthermore, we end up doing some
calculations twice without any guarantee that the prefetch can offset
the cost.
It's not strong enough of an opinion to outright NAK it but I'm not
ACKing it either.
--
Mel Gorman
SUSE Labs
--
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:[~2018-01-24 16:43 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-01-24 2:30 [PATCH 1/2] free_pcppages_bulk: do not hold lock when picking pages to free Aaron Lu
2018-01-24 2:30 ` [PATCH 2/2] free_pcppages_bulk: prefetch buddy while not holding lock Aaron Lu
2018-01-24 16:43 ` Mel Gorman [this message]
2018-01-24 16:57 ` Dave Hansen
2018-01-24 18:19 ` Mel Gorman
2018-01-24 19:23 ` Dave Hansen
2018-01-24 21:12 ` Mel Gorman
2018-01-25 7:25 ` [PATCH v2 " Aaron Lu
2018-01-24 16:40 ` [PATCH 1/2] free_pcppages_bulk: do not hold lock when picking pages to free Mel Gorman
2018-01-25 7:21 ` [PATCH v2 " Aaron Lu
2018-02-15 12:06 ` Mel Gorman
2018-02-23 1:37 ` Aaron Lu
2018-02-15 12:46 ` Matthew Wilcox
2018-02-15 14:55 ` Mel Gorman
2018-02-23 1:42 ` Aaron Lu
2018-02-05 5:30 ` RFC: eliminate zone->lock contention for will-it-scale/page_fault1 on big server Aaron Lu
2018-02-05 5:31 ` [RFC PATCH 1/2] __free_one_page: skip merge for order-0 page unless compaction is in progress Aaron Lu
2018-02-05 22:17 ` Dave Hansen
2018-02-05 5:32 ` [RFC PATCH 2/2] rmqueue_bulk: avoid touching page structures under zone->lock Aaron Lu
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=20180124164344.lca63gjn7mefuiac@techsingularity.net \
--to=mgorman@techsingularity.net \
--cc=aaron.lu@intel.com \
--cc=ak@linux.intel.com \
--cc=akpm@linux-foundation.org \
--cc=dave.hansen@intel.com \
--cc=kemi.wang@intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mhocko@suse.com \
--cc=tim.c.chen@linux.intel.com \
--cc=vbabka@suse.cz \
--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).