From: Ingo Molnar <mingo@kernel.org>
To: Tim Chen <tim.c.chen@linux.intel.com>
Cc: Ingo Molnar <mingo@elte.hu>,
Andrea Arcangeli <aarcange@redhat.com>,
Mel Gorman <mgorman@suse.de>, "Shi, Alex" <alex.shi@intel.com>,
Andi Kleen <andi@firstfloor.org>,
Andrew Morton <akpm@linux-foundation.org>,
Michel Lespinasse <walken@google.com>,
Davidlohr Bueso <davidlohr.bueso@hp.com>,
"Wilcox, Matthew R" <matthew.r.wilcox@intel.com>,
Dave Hansen <dave.hansen@intel.com>,
Peter Zijlstra <a.p.zijlstra@chello.nl>,
Rik van Riel <riel@redhat.com>,
linux-kernel@vger.kernel.org, linux-mm <linux-mm@kvack.org>
Subject: Re: Performance regression from switching lock to rw-sem for anon-vma tree
Date: Fri, 28 Jun 2013 11:38:09 +0200 [thread overview]
Message-ID: <20130628093809.GB29205@gmail.com> (raw)
In-Reply-To: <1372375873.22432.200.camel@schen9-DESK>
* Tim Chen <tim.c.chen@linux.intel.com> wrote:
> I tried some tweaking that checks sem->count for read owned lock. Even
> though it reduces the percentage of acquisitions that need sleeping by
> 8.14% (from 18.6% to 10.46%), it increases the writer acquisition
> blocked count by 11%. This change still doesn't boost throughput and has
> a tiny regression for the workload.
>
> Opt Spin Opt Spin
> (with tweak)
> Writer acquisition blocked count 7359040 8168006
> Blocked by reader 0.55% 0.52%
> Lock acquired first attempt (lock stealing) 16.92% 19.70%
> Lock acquired second attempt (1 sleep) 17.60% 9.32%
> Lock acquired after more than 1 sleep 1.00% 1.14%
> Lock acquired with optimistic spin 64.48% 69.84%
> Optimistic spin abort 1 11.77% 1.14%
> Optimistic spin abort 2 6.81% 9.22%
> Optimistic spin abort 3 0.02% 0.10%
So lock stealing+spinning now acquires the lock successfully ~90% of the
time, the remaining sleeps are:
> Lock acquired second attempt (1 sleep) ...... 9.32%
And the reason these sleeps are mostly due to:
> Optimistic spin abort 2 ..... 9.22%
Right?
So this particular #2 abort point is:
| preempt_disable();
| for (;;) {
| owner = ACCESS_ONCE(sem->owner);
| if (owner && !rwsem_spin_on_owner(sem, owner))
| break; <--------------------------- abort (2)
Next step would be to investigate why we decide to not spin there, why
does rwsem_spin_on_owner() fail?
If I got all the patches right, rwsem_spin_on_owner() is this:
+static noinline
+int rwsem_spin_on_owner(struct rw_semaphore *lock, struct task_struct *owner)
+{
+ rcu_read_lock();
+ while (owner_running(lock, owner)) {
+ if (need_resched())
+ break;
+
+ arch_mutex_cpu_relax();
+ }
+ rcu_read_unlock();
+
+ /*
+ * We break out the loop above on need_resched() and when the
+ * owner changed, which is a sign for heavy contention. Return
+ * success only when lock->owner is NULL.
+ */
+ return lock->owner == NULL;
+}
where owner_running() is similar to the mutex spinning code: it in the end
checks owner->on_cpu - like the mutex code.
If my analysis is correct so far then it might be useful to add two more
stats: did rwsem_spin_on_owner() fail because lock->owner == NULL [owner
released the rwsem], or because owner_running() failed [owner went to
sleep]?
Thanks,
Ingo
next prev parent reply other threads:[~2013-06-28 9:38 UTC|newest]
Thread overview: 46+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-06-13 23:26 Performance regression from switching lock to rw-sem for anon-vma tree Tim Chen
2013-06-19 13:16 ` Ingo Molnar
2013-06-19 16:53 ` Tim Chen
2013-06-26 0:19 ` Tim Chen
2013-06-26 9:51 ` Ingo Molnar
2013-06-26 21:36 ` Tim Chen
2013-06-27 0:25 ` Tim Chen
2013-06-27 8:36 ` Ingo Molnar
2013-06-27 20:53 ` Tim Chen
2013-06-27 23:31 ` Tim Chen
2013-06-28 9:38 ` Ingo Molnar [this message]
2013-06-28 21:04 ` Tim Chen
2013-06-29 7:12 ` Ingo Molnar
2013-07-01 20:28 ` Tim Chen
2013-07-02 6:45 ` Ingo Molnar
2013-07-16 17:53 ` Tim Chen
2013-07-23 9:45 ` Ingo Molnar
2013-07-23 9:51 ` Peter Zijlstra
2013-07-23 9:53 ` Ingo Molnar
2013-07-30 0:13 ` Tim Chen
2013-07-30 19:24 ` Ingo Molnar
2013-08-05 22:08 ` Tim Chen
2013-07-30 19:59 ` Davidlohr Bueso
2013-07-30 20:34 ` Tim Chen
2013-07-30 21:45 ` Davidlohr Bueso
2013-08-06 23:55 ` Davidlohr Bueso
2013-08-07 0:56 ` Tim Chen
2013-08-12 18:52 ` Ingo Molnar
2013-08-12 20:10 ` Tim Chen
2013-06-28 9:20 ` Ingo Molnar
[not found] <1371165333.27102.568.camel@schen9-DESK>
[not found] ` <1371167015.1754.14.camel@buesod1.americas.hpqcorp.net>
2013-06-14 16:09 ` Tim Chen
2013-06-14 22:31 ` Davidlohr Bueso
2013-06-14 22:44 ` Tim Chen
2013-06-14 22:47 ` Michel Lespinasse
2013-06-17 22:27 ` Tim Chen
2013-06-16 9:50 ` Alex Shi
2013-06-17 16:22 ` Davidlohr Bueso
2013-06-17 18:45 ` Tim Chen
2013-06-17 19:05 ` Davidlohr Bueso
2013-06-17 22:28 ` Tim Chen
2013-06-17 23:18 ` Alex Shi
2013-06-17 23:20 ` Alex Shi
2013-06-17 23:35 ` Davidlohr Bueso
2013-06-18 0:08 ` Tim Chen
2013-06-19 23:11 ` Davidlohr Bueso
2013-06-19 23:24 ` Tim Chen
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=20130628093809.GB29205@gmail.com \
--to=mingo@kernel.org \
--cc=a.p.zijlstra@chello.nl \
--cc=aarcange@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=alex.shi@intel.com \
--cc=andi@firstfloor.org \
--cc=dave.hansen@intel.com \
--cc=davidlohr.bueso@hp.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=matthew.r.wilcox@intel.com \
--cc=mgorman@suse.de \
--cc=mingo@elte.hu \
--cc=riel@redhat.com \
--cc=tim.c.chen@linux.intel.com \
--cc=walken@google.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).