From: Gleb Natapov <gleb@redhat.com>
To: Pekka Enberg <penberg@cs.helsinki.fi>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>,
linux-mm@kvack.org, kosaki.motohiro@jp.fujitsu.com,
linux-kernel@vger.kernel.org, linux-api@vger.kernel.org,
akpm@linux-foundation.org, andrew.c.morrow@gmail.com,
"Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Subject: Re: [PATCH v6] add MAP_UNLOCKED mmap flag
Date: Tue, 19 Jan 2010 10:26:38 +0200 [thread overview]
Message-ID: <20100119082638.GK14345@redhat.com> (raw)
In-Reply-To: <84144f021001190007q54a334dfwed64189e6cf0b7c4@mail.gmail.com>
On Tue, Jan 19, 2010 at 10:07:07AM +0200, Pekka Enberg wrote:
> Hi Gleb,
>
> On Tue, Jan 19, 2010 at 9:52 AM, Gleb Natapov <gleb@redhat.com> wrote:
> >> It would be probably useful if you could point us to the application
> >> source code that actually wants this feature.
> >>
> > This is two line patch to qemu that calls mlockall(MCL_CURRENT|MCL_FUTURE)
> > at the beginning of the main() and changes guest memory allocation to
> > use MAP_UNLOCKED flag. All alternative solutions in this thread suggest
> > that I should rewrite qemu + all library it uses. You see why I can't
> > take them seriously?
>
> Well, that's not going to be portable, is it, so the application
KVM is not portable ;) and that is what my main interest is.
> design would still be broken, no? Did you try using (or extending)
> posix_madvise(MADV_DONTNEED) for the guest address space? It seems to
After mlockall() I can't even allocate guest address space. Or do you mean
instead of mlockall()? Then how MADV_DONTNEED will help? It just drops
page table for the address range (which is not what I need) and does not
have any long time effect.
> me that you're trying to use a big hammer (mlock) when a polite hint
> for the VM would probably be sufficient for it do its job.
>
I what to tell to VM "swap this, don't swap that" and as far as I see
there is no other way to do it currently.
--
Gleb.
--
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>
WARNING: multiple messages have this Message-ID (diff)
From: Gleb Natapov <gleb@redhat.com>
To: Pekka Enberg <penberg@cs.helsinki.fi>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>,
linux-mm@kvack.org, kosaki.motohiro@jp.fujitsu.com,
linux-kernel@vger.kernel.org, linux-api@vger.kernel.org,
akpm@linux-foundation.org, andrew.c.morrow@gmail.com,
"Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Subject: Re: [PATCH v6] add MAP_UNLOCKED mmap flag
Date: Tue, 19 Jan 2010 10:26:38 +0200 [thread overview]
Message-ID: <20100119082638.GK14345@redhat.com> (raw)
In-Reply-To: <84144f021001190007q54a334dfwed64189e6cf0b7c4@mail.gmail.com>
On Tue, Jan 19, 2010 at 10:07:07AM +0200, Pekka Enberg wrote:
> Hi Gleb,
>
> On Tue, Jan 19, 2010 at 9:52 AM, Gleb Natapov <gleb@redhat.com> wrote:
> >> It would be probably useful if you could point us to the application
> >> source code that actually wants this feature.
> >>
> > This is two line patch to qemu that calls mlockall(MCL_CURRENT|MCL_FUTURE)
> > at the beginning of the main() and changes guest memory allocation to
> > use MAP_UNLOCKED flag. All alternative solutions in this thread suggest
> > that I should rewrite qemu + all library it uses. You see why I can't
> > take them seriously?
>
> Well, that's not going to be portable, is it, so the application
KVM is not portable ;) and that is what my main interest is.
> design would still be broken, no? Did you try using (or extending)
> posix_madvise(MADV_DONTNEED) for the guest address space? It seems to
After mlockall() I can't even allocate guest address space. Or do you mean
instead of mlockall()? Then how MADV_DONTNEED will help? It just drops
page table for the address range (which is not what I need) and does not
have any long time effect.
> me that you're trying to use a big hammer (mlock) when a polite hint
> for the VM would probably be sufficient for it do its job.
>
I what to tell to VM "swap this, don't swap that" and as far as I see
there is no other way to do it currently.
--
Gleb.
next prev parent reply other threads:[~2010-01-19 8:26 UTC|newest]
Thread overview: 82+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-01-18 13:37 [PATCH v6] add MAP_UNLOCKED mmap flag Gleb Natapov
2010-01-18 13:37 ` Gleb Natapov
2010-01-18 13:37 ` Gleb Natapov
[not found] ` <20100118133755.GG30698-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2010-01-18 14:09 ` Pekka Enberg
2010-01-18 14:09 ` Pekka Enberg
2010-01-18 14:09 ` Pekka Enberg
2010-01-18 14:19 ` Gleb Natapov
2010-01-18 14:19 ` Gleb Natapov
2010-01-18 14:32 ` Alan Cox
2010-01-18 14:32 ` Alan Cox
2010-01-18 14:35 ` Gleb Natapov
2010-01-18 14:35 ` Gleb Natapov
2010-01-18 14:49 ` Peter Zijlstra
2010-01-18 14:49 ` Peter Zijlstra
2010-01-18 15:01 ` Gleb Natapov
2010-01-18 15:01 ` Gleb Natapov
2010-01-18 15:06 ` Peter Zijlstra
2010-01-18 15:06 ` Peter Zijlstra
2010-01-18 15:11 ` Avi Kivity
2010-01-18 15:11 ` Avi Kivity
2010-01-18 15:14 ` Peter Zijlstra
2010-01-18 15:14 ` Peter Zijlstra
2010-01-18 15:19 ` Avi Kivity
2010-01-18 15:19 ` Avi Kivity
[not found] ` <4B547C09.8010906-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2010-01-18 15:24 ` Peter Zijlstra
2010-01-18 15:24 ` Peter Zijlstra
2010-01-18 15:24 ` Peter Zijlstra
2010-01-18 15:41 ` Alan Cox
2010-01-18 15:41 ` Alan Cox
2010-01-18 15:44 ` Peter Zijlstra
2010-01-18 15:44 ` Peter Zijlstra
2010-01-18 17:11 ` Gleb Natapov
2010-01-18 17:11 ` Gleb Natapov
2010-01-18 15:14 ` Gleb Natapov
2010-01-18 15:14 ` Gleb Natapov
2010-01-19 0:12 ` KOSAKI Motohiro
2010-01-19 0:12 ` KOSAKI Motohiro
2010-01-18 16:05 ` Pekka Enberg
2010-01-18 16:05 ` Pekka Enberg
2010-01-18 17:08 ` Gleb Natapov
2010-01-18 17:08 ` Gleb Natapov
2010-01-18 18:09 ` Pekka Enberg
2010-01-18 18:09 ` Pekka Enberg
2010-01-18 18:19 ` Gleb Natapov
2010-01-18 18:19 ` Gleb Natapov
2010-01-18 19:10 ` Alan Cox
2010-01-18 19:10 ` Alan Cox
2010-01-19 7:17 ` Gleb Natapov
2010-01-19 7:17 ` Gleb Natapov
2010-01-19 7:37 ` Pekka Enberg
2010-01-19 7:37 ` Pekka Enberg
2010-01-19 7:52 ` Gleb Natapov
2010-01-19 7:52 ` Gleb Natapov
[not found] ` <20100119075205.GI14345-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2010-01-19 8:07 ` Pekka Enberg
2010-01-19 8:07 ` Pekka Enberg
2010-01-19 8:07 ` Pekka Enberg
2010-01-19 8:26 ` Gleb Natapov [this message]
2010-01-19 8:26 ` Gleb Natapov
[not found] ` <20100119082638.GK14345-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2010-01-19 8:44 ` Pekka Enberg
2010-01-19 8:44 ` Pekka Enberg
2010-01-19 8:44 ` Pekka Enberg
2010-01-19 10:40 ` Gleb Natapov
2010-01-19 10:40 ` Gleb Natapov
[not found] ` <84144f021001190044s397c6665qb00af48235d2d818-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2010-01-19 12:48 ` Minchan Kim
2010-01-19 12:48 ` Minchan Kim
2010-01-19 12:48 ` Minchan Kim
2010-01-19 13:18 ` Pekka Enberg
2010-01-19 13:18 ` Pekka Enberg
2010-01-19 13:26 ` Gleb Natapov
2010-01-19 13:26 ` Gleb Natapov
2010-01-20 0:24 ` KOSAKI Motohiro
2010-01-20 0:24 ` KOSAKI Motohiro
2010-01-19 11:54 ` Alan Cox
2010-01-19 11:54 ` Alan Cox
2010-01-19 12:07 ` Gleb Natapov
2010-01-19 12:07 ` Gleb Natapov
2010-01-19 13:21 ` Alan Cox
2010-01-19 13:21 ` Alan Cox
2010-01-19 14:07 ` Minchan Kim
2010-01-19 14:07 ` Minchan Kim
2010-01-19 14:14 ` Gleb Natapov
2010-01-19 14:14 ` Gleb Natapov
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=20100119082638.GK14345@redhat.com \
--to=gleb@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=alan@lxorguk.ukuu.org.uk \
--cc=andrew.c.morrow@gmail.com \
--cc=kosaki.motohiro@jp.fujitsu.com \
--cc=linux-api@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=paulmck@linux.vnet.ibm.com \
--cc=penberg@cs.helsinki.fi \
/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.