From: Andi Kleen <andi@firstfloor.org>
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ingo Molnar <mingo@kernel.org>, Waiman Long <Waiman.Long@hp.com>,
Ingo Molnar <mingo@elte.hu>,
Andrew Morton <akpm@linux-foundation.org>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
Rik van Riel <riel@redhat.com>,
Peter Hurley <peter@hurleysoftware.com>,
Davidlohr Bueso <davidlohr.bueso@hp.com>,
Alex Shi <alex.shi@intel.com>,
Tim Chen <tim.c.chen@linux.intel.com>,
Peter Zijlstra <a.p.zijlstra@chello.nl>,
Andrea Arcangeli <aarcange@redhat.com>,
Matthew R Wilcox <matthew.r.wilcox@intel.com>,
Dave Hansen <dave.hansen@intel.com>,
Michel Lespinasse <walken@google.com>,
Andi Kleen <andi@firstfloor.org>,
"Chandramouleeswaran, Aswin" <aswin@hp.com>,
"Norton, Scott J" <scott.norton@hp.com>
Subject: Re: [PATCH] rwsem: reduce spinlock contention in wakeup code path
Date: Sat, 28 Sep 2013 21:13:39 +0200 [thread overview]
Message-ID: <20130928191339.GP18242@two.firstfloor.org> (raw)
In-Reply-To: <CA+55aFyx-Lpqd8i2tHvhXCqL+nJZPq-6SikEkb-cQZEU9ogRjA@mail.gmail.com>
> Of course, since then, we may well have screwed things up and now we
> sleep under it, but I still really think it was a mistake to do it in
> the first place.
>
> So if the primary reason for this is really just that f*cking anon_vma
> lock, then I would seriously suggest:
>
> - turn it back into a spinlock (or rwlock_t, since we subsequently
> separated the read and write paths)
Yes please. spinlocks/rwlocks have so much nicer performance behavior than
rwsems/mutexes (which noone seems to fully understand)
We had also significant performance regressions from every such
spinning->sleeping change in the VM (this was just the latest)
And afaik anon_vma is usually hold short.
-Andi
--
ak@linux.intel.com -- Speaking for myself only.
next prev parent reply other threads:[~2013-09-28 19:14 UTC|newest]
Thread overview: 55+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-09-27 19:00 [PATCH] rwsem: reduce spinlock contention in wakeup code path Waiman Long
2013-09-27 19:28 ` Linus Torvalds
2013-09-27 19:39 ` Davidlohr Bueso
2013-09-27 21:49 ` Tim Chen
2013-09-28 6:45 ` Ingo Molnar
2013-09-28 7:41 ` Ingo Molnar
2013-09-28 18:55 ` Linus Torvalds
2013-09-28 19:13 ` Andi Kleen [this message]
2013-09-28 19:22 ` Linus Torvalds
2013-09-28 19:21 ` Ingo Molnar
2013-09-28 19:33 ` Linus Torvalds
2013-09-28 19:39 ` Ingo Molnar
2013-09-30 10:44 ` Peter Zijlstra
2013-09-30 16:13 ` Linus Torvalds
2013-09-30 16:41 ` Peter Zijlstra
2013-10-01 7:28 ` Ingo Molnar
2013-10-01 8:09 ` Peter Zijlstra
2013-10-01 8:25 ` Ingo Molnar
2013-09-30 15:58 ` Waiman Long
2013-10-01 7:33 ` Ingo Molnar
2013-10-01 20:03 ` Waiman Long
2013-09-28 19:37 ` [PATCH] anon_vmas: Convert the rwsem to an rwlock_t Ingo Molnar
2013-09-28 19:43 ` Linus Torvalds
2013-09-28 19:46 ` Ingo Molnar
2013-09-28 19:52 ` [PATCH, v2] " Ingo Molnar
2013-09-30 11:00 ` Peter Zijlstra
2013-09-30 11:44 ` Peter Zijlstra
2013-09-30 17:03 ` Andrew Morton
2013-09-30 17:25 ` Linus Torvalds
2013-09-30 17:10 ` Tim Chen
2013-09-30 18:14 ` Peter Zijlstra
2013-09-30 19:23 ` Tim Chen
2013-09-30 19:35 ` Waiman Long
2013-09-30 19:47 ` Tim Chen
2013-09-30 22:03 ` Tim Chen
2013-10-01 2:41 ` Waiman Long
2013-09-30 8:52 ` [PATCH] " Andrea Arcangeli
2013-09-30 14:40 ` Jerome Glisse
2013-09-30 16:26 ` Linus Torvalds
2013-09-30 19:16 ` Andrea Arcangeli
2013-09-30 18:21 ` Andi Kleen
2013-09-30 19:04 ` Jerome Glisse
2013-09-29 23:06 ` [PATCH] rwsem: reduce spinlock contention in wakeup code path Davidlohr Bueso
2013-09-29 23:26 ` Linus Torvalds
2013-09-30 0:40 ` Davidlohr Bueso
2013-09-30 0:51 ` Linus Torvalds
2013-09-30 6:57 ` Ingo Molnar
2013-09-30 1:11 ` Michel Lespinasse
2013-09-30 7:05 ` Ingo Molnar
2013-09-30 16:03 ` Waiman Long
2013-09-30 10:46 ` Peter Zijlstra
2013-10-01 7:48 ` Ingo Molnar
2013-10-01 8:10 ` Peter Zijlstra
2013-09-27 19:32 ` Peter Hurley
2013-09-28 0:46 ` Waiman Long
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=20130928191339.GP18242@two.firstfloor.org \
--to=andi@firstfloor.org \
--cc=Waiman.Long@hp.com \
--cc=a.p.zijlstra@chello.nl \
--cc=aarcange@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=alex.shi@intel.com \
--cc=aswin@hp.com \
--cc=dave.hansen@intel.com \
--cc=davidlohr.bueso@hp.com \
--cc=linux-kernel@vger.kernel.org \
--cc=matthew.r.wilcox@intel.com \
--cc=mingo@elte.hu \
--cc=mingo@kernel.org \
--cc=peter@hurleysoftware.com \
--cc=riel@redhat.com \
--cc=scott.norton@hp.com \
--cc=tim.c.chen@linux.intel.com \
--cc=torvalds@linux-foundation.org \
--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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.