linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Benjamin LaHaise <bcrl@redhat.com>
To: Andrea Arcangeli <andrea@suse.de>
Cc: Andrew Morton <akpm@digeo.com>,
	mingo@elte.hu, hugh@veritas.com, dmccr@us.ibm.com,
	linux-kernel@vger.kernel.org, linux-mm@kvack.org
Subject: Re: objrmap and vmtruncate
Date: Fri, 4 Apr 2003 20:52:48 -0500	[thread overview]
Message-ID: <20030404205248.C21819@redhat.com> (raw)
In-Reply-To: <20030405013143.GJ16293@dualathlon.random>; from andrea@suse.de on Sat, Apr 05, 2003 at 03:31:43AM +0200

On Sat, Apr 05, 2003 at 03:31:43AM +0200, Andrea Arcangeli wrote:
> Also consider this significant factor: the larger the shmfs the smaller
> the nonlinear 1G window will be and the higher the trashing. With 32G of
> bigpages the remap_file_pages will trash like crazy generating an order
> of mangnitude more of "window misses". I mean 32bit are just pushed at
> the limit today regardless the lack of remap_file_pages. Example, if
> you don't use largepages going past 16G of shm is going to be derimental.
> The cost of the mmap doesn't sounds like the showstopper.

You're guessing here.  At least for oracle, that behaviour is dependant on 
the locality of accesses.  Given that each user has their own process you 
can bet there is a fair amount of locality to their transactions.

> you could try to avoid the need of the sysctl by teaching the vm to
> unmap such vma, but I don't think it worth and I'm sure those apps
> prefers to have the stuff pinned anyways w/o the risk of sigbus and w/o
> the need of mlock and it looks cleaner to me to avoid any mess with the
> vm and long term nobody will care about this sysctl since 64bit will run
> so much fatster w/o any remap_file_pages and tlb flush running at all

It is still useful for things outside of the pure databases on 32 bits 
realm.  Consider a fast bochs running 32 bit apps on a 64 bit machine -- 
should it have to deal with the overhead of zillions of vmas for emulating 
page tables?

If anything, I think we should be moving in the direction of doing more 
along the lines of remap_file_pages: things like executables might as well 
keep their state in page tables since we never discard them and instead 
toss the vma out the window.

		-ben
-- 
Junk email?  <a href="mailto:aart@kvack.org">aart@kvack.org</a>
--
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:"aart@kvack.org">aart@kvack.org</a>

  reply	other threads:[~2003-04-05  1:52 UTC|newest]

Thread overview: 105+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2003-04-04 14:34 Hugh Dickins
2003-04-04 16:14 ` William Lee Irwin III
2003-04-04 16:29   ` Hugh Dickins
2003-04-04 18:54 ` Andrew Morton
2003-04-04 21:43   ` Hugh Dickins
2003-04-04 21:45   ` Andrea Arcangeli
2003-04-04 21:58     ` Benjamin LaHaise
2003-04-04 23:07     ` Andrew Morton
2003-04-05  0:03       ` Andrea Arcangeli
2003-04-05  0:31         ` Andrew Morton
2003-04-05  1:31           ` Andrea Arcangeli
2003-04-05  1:52             ` Benjamin LaHaise [this message]
2003-04-05  2:22               ` Andrea Arcangeli
2003-04-05 10:01                 ` Jamie Lokier
2003-04-05 10:11                   ` William Lee Irwin III
2003-04-05  2:06             ` Andrew Morton
2003-04-05  2:24               ` Andrea Arcangeli
2003-04-05  2:13           ` Martin J. Bligh
2003-04-05  2:44             ` Andrea Arcangeli
2003-04-05  3:24               ` Andrew Morton
2003-04-05 12:06                 ` Andrew Morton
2003-04-05 15:11                   ` Martin J. Bligh
     [not found]                     ` <20030405161758.1ee19bfa.akpm@digeo.com>
2003-04-06  0:17                       ` Andrew Morton
2003-04-06  7:07                       ` William Lee Irwin III
2003-04-05 16:30                   ` Andrea Arcangeli
2003-04-05 19:01                     ` Andrea Arcangeli
2003-04-05 20:14                       ` Andrew Morton
2003-04-05 21:24                     ` Andrew Morton
2003-04-05 22:06                       ` Andrea Arcangeli
2003-04-05 22:31                         ` Andrew Morton
2003-04-05 23:10                           ` Andrea Arcangeli
2003-04-06  1:58                             ` Andrew Morton
2003-04-06 14:47                               ` Andrea Arcangeli
2003-04-06 21:35                                 ` William Lee Irwin III
2003-04-06  7:38                             ` William Lee Irwin III
2003-04-06 14:51                               ` Andrea Arcangeli
2003-04-06 12:37                           ` Jamie Lokier
2003-04-06 13:12                             ` William Lee Irwin III
2003-04-22 11:00                           ` Ingo Molnar
2003-04-22 11:54                             ` William Lee Irwin III
2003-04-22 14:31                               ` Ingo Molnar
2003-04-22 14:56                                 ` William Lee Irwin III
2003-04-22 15:26                                   ` Ingo Molnar
2003-04-22 16:20                                     ` William Lee Irwin III
2003-04-22 16:57                                       ` Andrea Arcangeli
2003-04-22 17:21                                         ` William Lee Irwin III
2003-04-22 18:08                                           ` Andrea Arcangeli
2003-04-22 17:34                                         ` Ingo Molnar
2003-04-22 18:04                                           ` Benjamin LaHaise
2003-04-22 16:58                                       ` Martin J. Bligh
2003-04-22 12:37                             ` Andrea Arcangeli
2003-04-22 13:20                               ` William Lee Irwin III
2003-04-22 14:38                                 ` Martin J. Bligh
2003-04-22 15:10                                   ` William Lee Irwin III
2003-04-22 15:53                                     ` Martin J. Bligh
2003-04-22 14:52                                 ` Andrea Arcangeli
2003-04-22 14:29                             ` Martin J. Bligh
2003-04-22 15:07                               ` Ingo Molnar
2003-04-22 15:42                                 ` William Lee Irwin III
2003-04-22 15:55                                   ` Ingo Molnar
2003-04-22 16:58                                     ` William Lee Irwin III
2003-04-22 17:07                                       ` Ingo Molnar
2003-04-22 15:16                               ` Andrea Arcangeli
2003-04-22 15:49                               ` Ingo Molnar
2003-04-22 16:16                                 ` Martin J. Bligh
2003-04-22 17:24                                   ` Ingo Molnar
2003-04-22 17:45                                   ` John Bradford
2003-04-22 14:32                             ` Martin J. Bligh
2003-04-22 15:09                               ` Ingo Molnar
2003-04-05 21:34                     ` Rik van Riel
2003-04-06  9:29                     ` Benjamin LaHaise
2003-04-05 23:25                   ` William Lee Irwin III
2003-04-05 23:57                     ` Andrew Morton
2003-04-06  0:14                       ` Andrea Arcangeli
2003-04-06  1:39                         ` Andrew Morton
2003-04-06  2:13                       ` William Lee Irwin III
2003-04-06  9:26                     ` Benjamin LaHaise
2003-04-06  9:41                       ` William Lee Irwin III
2003-04-06  9:54                         ` William Lee Irwin III
2003-04-06  2:23                   ` Martin J. Bligh
2003-04-06  3:55                     ` Andrew Morton
2003-04-06  3:08                       ` Martin J. Bligh
2003-04-06  7:42                         ` William Lee Irwin III
2003-04-06 14:49                     ` Alan Cox
2003-04-06 16:13                       ` Martin J. Bligh
2003-04-06 21:34                         ` subobj-rmap Martin J. Bligh
2003-04-06 21:42                           ` subobj-rmap Rik van Riel
2003-04-06 21:55                             ` subobj-rmap Jamie Lokier
2003-04-06 22:39                               ` subobj-rmap William Lee Irwin III
2003-04-06 22:03                             ` subobj-rmap Martin J. Bligh
2003-04-06 22:06                               ` subobj-rmap Martin J. Bligh
2003-04-06 22:15                               ` subobj-rmap Andrea Arcangeli
2003-04-06 22:25                                 ` subobj-rmap Martin J. Bligh
2003-04-07 21:25                                   ` subobj-rmap Andrea Arcangeli
2003-04-06 23:06                               ` subobj-rmap Jamie Lokier
2003-04-06 23:26                                 ` subobj-rmap Martin J. Bligh
2003-04-05  3:45               ` objrmap and vmtruncate Martin J. Bligh
2003-04-05  3:59               ` Rik van Riel
2003-04-05  4:10                 ` William Lee Irwin III
2003-04-05  4:49                   ` Martin J. Bligh
2003-04-05 13:31                     ` Rik van Riel
2003-04-05  4:52               ` Martin J. Bligh
2003-04-05  3:22             ` Andrew Morton
2003-04-05  3:35               ` Martin J. Bligh
2003-04-05  3:53       ` Rik van Riel

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=20030404205248.C21819@redhat.com \
    --to=bcrl@redhat.com \
    --cc=akpm@digeo.com \
    --cc=andrea@suse.de \
    --cc=dmccr@us.ibm.com \
    --cc=hugh@veritas.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mingo@elte.hu \
    /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