From: Andrea Arcangeli <andrea@qumranet.com>
To: Christoph Lameter <clameter@sgi.com>
Cc: Hugh Dickins <hugh@veritas.com>, Robin Holt <holt@sgi.com>,
Avi Kivity <avi@qumranet.com>, Izik Eidus <izike@qumranet.com>,
kvm-devel@lists.sourceforge.net,
Peter Zijlstra <a.p.zijlstra@chello.nl>,
general@lists.openfabrics.org,
Steve Wise <swise@opengridcomputing.com>,
Roland Dreier <rdreier@cisco.com>,
Kanoj Sarcar <kanojsarcar@yahoo.com>,
steiner@sgi.com, linux-kernel@vger.kernel.org,
linux-mm@kvack.org, daniel.blueman@quadrics.com
Subject: Re: [patch 5/9] Convert anon_vma lock to rw_sem and refcount
Date: Thu, 3 Apr 2008 00:12:28 +0200 [thread overview]
Message-ID: <20080402221228.GX19189@duo.random> (raw)
In-Reply-To: <Pine.LNX.4.64.0804021455180.31247@schroedinger.engr.sgi.com>
On Wed, Apr 02, 2008 at 02:56:25PM -0700, Christoph Lameter wrote:
> I am a bit surprised that brk performance is that important. There may be
I think it's not brk but fork that is being slowed down, did you
oprofile? AIM forks a lot... The write side fast path generating the
overscheduling I guess is when the new vmas are created for the child
and queued in the parent anon-vma in O(1), so immediate, even
preempt-rt would be ok with it spinning and not scheduling, it's just
a list_add (much faster than schedule() indeed). Every time there's a
collision when multiple child forks simultaneously and they all try to
queue in the same anon-vma, things will slowdown.
WARNING: multiple messages have this Message-ID (diff)
From: Andrea Arcangeli <andrea@qumranet.com>
To: Christoph Lameter <clameter@sgi.com>
Cc: steiner@sgi.com, Peter Zijlstra <a.p.zijlstra@chello.nl>,
linux-mm@kvack.org, Izik Eidus <izike@qumranet.com>,
Kanoj Sarcar <kanojsarcar@yahoo.com>,
Roland Dreier <rdreier@cisco.com>,
linux-kernel@vger.kernel.org, Avi Kivity <avi@qumranet.com>,
kvm-devel@lists.sourceforge.net, daniel.blueman@quadrics.com,
Robin Holt <holt@sgi.com>,
general@lists.openfabrics.org, Hugh Dickins <hugh@veritas.com>
Subject: [ofa-general] Re: [patch 5/9] Convert anon_vma lock to rw_sem and refcount
Date: Thu, 3 Apr 2008 00:12:28 +0200 [thread overview]
Message-ID: <20080402221228.GX19189@duo.random> (raw)
In-Reply-To: <Pine.LNX.4.64.0804021455180.31247@schroedinger.engr.sgi.com>
On Wed, Apr 02, 2008 at 02:56:25PM -0700, Christoph Lameter wrote:
> I am a bit surprised that brk performance is that important. There may be
I think it's not brk but fork that is being slowed down, did you
oprofile? AIM forks a lot... The write side fast path generating the
overscheduling I guess is when the new vmas are created for the child
and queued in the parent anon-vma in O(1), so immediate, even
preempt-rt would be ok with it spinning and not scheduling, it's just
a list_add (much faster than schedule() indeed). Every time there's a
collision when multiple child forks simultaneously and they all try to
queue in the same anon-vma, things will slowdown.
next prev parent reply other threads:[~2008-04-02 22:12 UTC|newest]
Thread overview: 93+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-04-01 20:55 [patch 0/9] [RFC] EMM Notifier V2 Christoph Lameter
2008-04-01 20:55 ` [ofa-general] " Christoph Lameter
2008-04-01 20:55 ` [patch 1/9] EMM Notifier: The notifier calls Christoph Lameter
2008-04-01 20:55 ` Christoph Lameter
2008-04-01 21:14 ` Peter Zijlstra
2008-04-01 21:38 ` Paul E. McKenney
2008-04-02 17:44 ` Christoph Lameter
2008-04-02 18:43 ` EMM: Fix rcu handling and spelling Christoph Lameter
2008-04-02 19:02 ` Paul E. McKenney
2008-04-02 6:49 ` [patch 1/9] EMM Notifier: The notifier calls Andrea Arcangeli
2008-04-02 6:49 ` [ofa-general] " Andrea Arcangeli
2008-04-02 10:59 ` Robin Holt
2008-04-02 10:59 ` [ofa-general] " Robin Holt
2008-04-02 11:16 ` Andrea Arcangeli
2008-04-02 11:16 ` Andrea Arcangeli
2008-04-02 14:26 ` Robin Holt
2008-04-02 17:59 ` Christoph Lameter
2008-04-02 17:59 ` [ofa-general] " Christoph Lameter
2008-04-02 19:03 ` EMM: Fixup return value handling of emm_notify() Christoph Lameter
2008-04-02 19:03 ` [ofa-general] " Christoph Lameter
2008-04-02 21:25 ` Andrea Arcangeli
2008-04-02 21:25 ` [ofa-general] " Andrea Arcangeli
2008-04-02 21:33 ` Christoph Lameter
2008-04-02 21:33 ` [ofa-general] " Christoph Lameter
2008-04-03 10:40 ` Peter Zijlstra
2008-04-03 10:40 ` [ofa-general] " Peter Zijlstra
2008-04-03 15:00 ` Andrea Arcangeli
2008-04-03 15:00 ` [ofa-general] " Andrea Arcangeli
2008-04-03 19:14 ` Christoph Lameter
2008-04-03 19:14 ` [ofa-general] " Christoph Lameter
2008-04-02 21:05 ` EMM: Require single threadedness for registration Christoph Lameter
2008-04-02 21:05 ` [ofa-general] " Christoph Lameter
2008-04-02 22:01 ` Andrea Arcangeli
2008-04-02 22:01 ` [ofa-general] " Andrea Arcangeli
2008-04-02 22:06 ` Christoph Lameter
2008-04-02 22:06 ` [ofa-general] " Christoph Lameter
2008-04-02 22:17 ` Andrea Arcangeli
2008-04-02 22:17 ` [ofa-general] " Andrea Arcangeli
2008-04-02 22:41 ` Christoph Lameter
2008-04-02 22:41 ` [ofa-general] " Christoph Lameter
2008-04-03 1:24 ` EMM: disable other notifiers before register and unregister Christoph Lameter
2008-04-03 1:24 ` Christoph Lameter
2008-04-03 10:40 ` Peter Zijlstra
2008-04-03 10:40 ` [ofa-general] " Peter Zijlstra
2008-04-03 15:29 ` Andrea Arcangeli
2008-04-03 19:20 ` Christoph Lameter
2008-04-03 19:20 ` [ofa-general] " Christoph Lameter
2008-04-03 20:23 ` Christoph Lameter
2008-04-04 12:30 ` Andrea Arcangeli
2008-04-04 12:30 ` Andrea Arcangeli
2008-04-04 20:20 ` [PATCH] mmu notifier #v11 Andrea Arcangeli
2008-04-04 20:20 ` [ofa-general] " Andrea Arcangeli
2008-04-04 22:06 ` Christoph Lameter
2008-04-05 0:23 ` Andrea Arcangeli
2008-04-05 0:23 ` [ofa-general] " Andrea Arcangeli
2008-04-07 5:45 ` Christoph Lameter
2008-04-07 5:45 ` Christoph Lameter
2008-04-07 6:02 ` Andrea Arcangeli
2008-04-07 6:02 ` [ofa-general] " Andrea Arcangeli
2008-04-02 21:53 ` [patch 1/9] EMM Notifier: The notifier calls Andrea Arcangeli
2008-04-02 21:53 ` [ofa-general] " Andrea Arcangeli
2008-04-02 21:54 ` Christoph Lameter
2008-04-02 21:54 ` [ofa-general] " Christoph Lameter
2008-04-02 22:09 ` Andrea Arcangeli
2008-04-02 22:09 ` [ofa-general] " Andrea Arcangeli
2008-04-02 23:04 ` Christoph Lameter
2008-04-02 23:04 ` [ofa-general] " Christoph Lameter
2008-04-01 20:55 ` [patch 2/9] Move tlb flushing into free_pgtables Christoph Lameter
2008-04-01 20:55 ` [ofa-general] " Christoph Lameter
2008-04-01 20:55 ` [patch 3/9] Convert i_mmap_lock to i_mmap_sem Christoph Lameter
2008-04-01 20:55 ` [ofa-general] " Christoph Lameter
2008-04-01 20:55 ` [patch 4/9] Remove tlb pointer from the parameters of unmap vmas Christoph Lameter
2008-04-01 20:55 ` Christoph Lameter
2008-04-01 20:55 ` [patch 5/9] Convert anon_vma lock to rw_sem and refcount Christoph Lameter
2008-04-01 20:55 ` Christoph Lameter
2008-04-02 17:50 ` Andrea Arcangeli
2008-04-02 17:50 ` Andrea Arcangeli
2008-04-02 18:15 ` Christoph Lameter
2008-04-02 18:15 ` Christoph Lameter
2008-04-02 21:56 ` Andrea Arcangeli
2008-04-02 21:56 ` [ofa-general] " Andrea Arcangeli
2008-04-02 21:56 ` Christoph Lameter
2008-04-02 21:56 ` [ofa-general] " Christoph Lameter
2008-04-02 22:12 ` Andrea Arcangeli [this message]
2008-04-02 22:12 ` Andrea Arcangeli
2008-04-01 20:55 ` [patch 6/9] This patch exports zap_page_range as it is needed by XPMEM Christoph Lameter
2008-04-01 20:55 ` Christoph Lameter
2008-04-01 20:55 ` [patch 7/9] Locking rules for taking multiple mmap_sem locks Christoph Lameter
2008-04-01 20:55 ` Christoph Lameter
2008-04-01 20:55 ` [patch 8/9] XPMEM: The device driver Christoph Lameter
2008-04-01 20:55 ` Christoph Lameter
2008-04-01 20:55 ` [patch 9/9] XPMEM: Simple example Christoph Lameter
2008-04-01 20:55 ` Christoph Lameter
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=20080402221228.GX19189@duo.random \
--to=andrea@qumranet.com \
--cc=a.p.zijlstra@chello.nl \
--cc=avi@qumranet.com \
--cc=clameter@sgi.com \
--cc=daniel.blueman@quadrics.com \
--cc=general@lists.openfabrics.org \
--cc=holt@sgi.com \
--cc=hugh@veritas.com \
--cc=izike@qumranet.com \
--cc=kanojsarcar@yahoo.com \
--cc=kvm-devel@lists.sourceforge.net \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=rdreier@cisco.com \
--cc=steiner@sgi.com \
--cc=swise@opengridcomputing.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.