From: Minchan Kim <minchan@kernel.org>
To: Waiman Long <longman@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>,
Vladimir Davydov <vdavydov.dev@gmail.com>,
Johannes Weiner <hannes@cmpxchg.org>,
Dave Chinner <david@fromorbit.com>,
linux-kernel@vger.kernel.org, linux-mm@kvack.org
Subject: Re: [PATCH] list_lru: Prefetch neighboring list entries before acquiring lock
Date: Fri, 1 Dec 2017 08:53:50 +0900 [thread overview]
Message-ID: <20171130235350.GA4389@bbox> (raw)
In-Reply-To: <414f9020-aba5-eef1-b689-36307dbdcfed@redhat.com>
On Thu, Nov 30, 2017 at 08:43:41AM -0500, Waiman Long wrote:
> On 11/29/2017 07:53 PM, Minchan Kim wrote:
> > Hello,
> >
> > On Wed, Nov 29, 2017 at 09:17:34AM -0500, Waiman Long wrote:
> >> The list_lru_del() function removes the given item from the LRU list.
> >> The operation looks simple, but it involves writing into the cachelines
> >> of the two neighboring list entries in order to get the deletion done.
> >> That can take a while if the cachelines aren't there yet, thus
> >> prolonging the lock hold time.
> >>
> >> To reduce the lock hold time, the cachelines of the two neighboring
> >> list entries are now prefetched before acquiring the list_lru_node's
> >> lock.
> >>
> >> Using a multi-threaded test program that created a large number
> >> of dentries and then killed them, the execution time was reduced
> >> from 38.5s to 36.6s after applying the patch on a 2-socket 36-core
> >> 72-thread x86-64 system.
> >>
> >> Signed-off-by: Waiman Long <longman@redhat.com>
> >> ---
> >> mm/list_lru.c | 10 +++++++++-
> >> 1 file changed, 9 insertions(+), 1 deletion(-)
> >>
> >> diff --git a/mm/list_lru.c b/mm/list_lru.c
> >> index f141f0c..65aae44 100644
> >> --- a/mm/list_lru.c
> >> +++ b/mm/list_lru.c
> >> @@ -132,8 +132,16 @@ bool list_lru_del(struct list_lru *lru, struct list_head *item)
> >> struct list_lru_node *nlru = &lru->node[nid];
> >> struct list_lru_one *l;
> >>
> >> + /*
> >> + * Prefetch the neighboring list entries to reduce lock hold time.
> >> + */
> >> + if (unlikely(list_empty(item)))
> >> + return false;
> >> + prefetchw(item->prev);
> >> + prefetchw(item->next);
> >> +
> > A question:
> >
> > A few month ago, I had a chance to measure prefetch effect with my testing
> > workload. For the clarification, it's not list_lru_del but list traverse
> > stuff so it might be similar.
> >
> > With my experiment at that time, it was really hard to find best place to
> > add prefetchw. Sometimes, it was too eariler or late so the effect was
> > not good, even worse on some cases.
> >
> > Also, the performance was different with each machine although my testing
> > machines was just two. ;-)
> >
> > So my question is what's a rule of thumb to add prefetch command?
> > Like your code, putting prefetch right before touching?
> >
> > I'm really wonder what's the rule to make every arch/machines happy
> > with prefetch.
>
> I add the prefetchw() before spin_lock() because the latency of the
> lockinig operation can be highly variable. There will have high latency
> when the lock is contended. With the prefetch, lock hold time will be
> reduced. In turn, it helps to reduce the amount of lock contention as
> well. If there is no lock contention, the prefetch won't help.
I knew it by your description. My point is prefetch optimization could
show different results by various architectures and workloads so
I wanted to know what kinds of rule we have to prove it's always win
or no harmful for *everycase* in geneal.
This is a performance patch and it's very micro-optimized topic so
I think we need more data to prove it. Maybe perf is best friend and need a
experiment with no lock contention case, at least.
Thanks.
--
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:[~2017-11-30 23:53 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-11-29 14:17 [PATCH] list_lru: Prefetch neighboring list entries before acquiring lock Waiman Long
2017-11-29 21:53 ` Andrew Morton
2017-11-30 0:42 ` Dave Chinner
2017-11-30 13:54 ` Waiman Long
2017-11-30 20:38 ` Dave Chinner
2017-11-30 20:55 ` Waiman Long
2017-11-30 20:47 ` Andrew Morton
2017-11-30 20:49 ` Waiman Long
2017-12-01 0:09 ` Minchan Kim
2017-12-01 14:14 ` Waiman Long
2017-12-01 22:02 ` Dave Chinner
2017-11-30 0:53 ` Minchan Kim
2017-11-30 13:43 ` Waiman Long
2017-11-30 23:53 ` Minchan Kim [this message]
2017-11-30 14:34 ` Matthew Wilcox
2017-12-05 14:49 ` Michal Hocko
2017-12-05 23:56 ` Andrew Morton
2017-12-06 8:07 ` 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=20171130235350.GA4389@bbox \
--to=minchan@kernel.org \
--cc=akpm@linux-foundation.org \
--cc=david@fromorbit.com \
--cc=hannes@cmpxchg.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=longman@redhat.com \
--cc=vdavydov.dev@gmail.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).