linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: glebn@voltaire.com (Gleb Natapov)
To: Andi Kleen <ak@suse.de>
Cc: Christoph Lameter <clameter@sgi.com>,
	Lee Schermerhorn <Lee.Schermerhorn@hp.com>,
	linux-mm <linux-mm@kvack.org>,
	Andrew Morton <akpm@linux-foundation.org>
Subject: Re: [PATCH] Document Linux Memory Policy
Date: Thu, 31 May 2007 14:04:12 +0300	[thread overview]
Message-ID: <20070531110412.GM4715@minantech.com> (raw)
In-Reply-To: <200705311243.20119.ak@suse.de>

On Thu, May 31, 2007 at 12:43:19PM +0200, Andi Kleen wrote:
> 
> > > The faulted page will use the memory policy of the task that faulted it 
> > > in. If that process has numa_set_localalloc() set then the page will be 
> > > located as closely as possible to the allocating thread.
> > 
> > Thanks. But I have to say this feels very unnatural.
> 
> What do you think is unnatural exactly? First one wins seems like a quite 
> natural policy to me.
No it is not (not always). I want to create shared memory for
interprocess communication. Process A will write into the memory and
process B will periodically poll it to see if there is a message there.
In NUMA system I want the physical memory for this VMA to be allocated
from node close to process B since it will use it much more frequently.
But I don't want to pre-fault all pages in process B to achieve this
because the region can be huge and because it doesn't guaranty much if
swapping is involved. So numa_set_localalloc() looks like it achieves
exactly this. Without this function I agree that the "first one wins" is
very sensible assumption, but when each process stated it's preferences
explicitly by calling the function it is not longer sensible to me as a
user of the API. When you start to thing about how memory policy may be
implemented in the kernel and understand that memory policy is a
property of an address space (is it?) and not a file then you start to
understand current behaviour, but this is implementation details.


> 
> > So to have 
> > desirable effect I have to create shared memory with shmget?
> 
> shmget behaves the same.
> 
Then I misinterpreted "Shared Policy" section from Lee's document.
It seems that he states that for memory region created with shmget the
policy is a property of a shared object and not of a process' address
space.

--
			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>

  reply	other threads:[~2007-05-31 11:04 UTC|newest]

Thread overview: 83+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-05-29 19:33 Lee Schermerhorn
2007-05-29 20:04 ` Christoph Lameter
2007-05-29 20:16   ` Andi Kleen
2007-05-30 16:17     ` Lee Schermerhorn
2007-05-30 17:41       ` Christoph Lameter
2007-05-31  8:20       ` Michael Kerrisk
2007-05-31 14:49         ` Lee Schermerhorn
2007-05-31 15:56           ` Michael Kerrisk
2007-06-01 21:15         ` [PATCH] enhance memory policy sys call man pages v1 Lee Schermerhorn
2007-07-23  6:11           ` Michael Kerrisk
2007-07-23  6:32           ` mbind.2 man page patch Michael Kerrisk
2007-07-23 14:26             ` Lee Schermerhorn
2007-07-26 17:19               ` Michael Kerrisk
2007-07-26 18:06                 ` Lee Schermerhorn
2007-07-26 18:18                   ` Michael Kerrisk
2007-07-23  6:32           ` get_mempolicy.2 " Michael Kerrisk
2007-07-28  9:31             ` Michael Kerrisk
2007-08-09 18:43               ` Lee Schermerhorn
2007-08-09 20:57                 ` Michael Kerrisk
2007-08-16 20:05               ` Andi Kleen
2007-08-18  5:50                 ` Michael Kerrisk
2007-08-21 15:45                   ` Lee Schermerhorn
2007-08-22  4:10                     ` Michael Kerrisk
2007-08-22 16:08                       ` [PATCH] Mempolicy Man Pages 2.64 1/3 - mbind.2 Lee Schermerhorn
2007-08-27 11:29                         ` Michael Kerrisk
2007-08-22 16:10                       ` [PATCH] Mempolicy Man Pages 2.64 2/3 - set_mempolicy.2 Lee Schermerhorn
2007-08-27 11:30                         ` Michael Kerrisk
2007-08-22 16:12                       ` [PATCH] Mempolicy Man Pages 2.64 3/3 - get_mempolicy.2 Lee Schermerhorn
2007-08-27 11:30                         ` Michael Kerrisk
2007-08-27 10:46                 ` get_mempolicy.2 man page patch Michael Kerrisk
2007-07-23  6:33           ` set_mempolicy.2 " Michael Kerrisk
2007-05-30 16:55   ` [PATCH] Document Linux Memory Policy Lee Schermerhorn
2007-05-30 17:56     ` Christoph Lameter
2007-05-31  6:18       ` Gleb Natapov
2007-05-31  6:41         ` Christoph Lameter
2007-05-31  6:47           ` Gleb Natapov
2007-05-31  6:56             ` Christoph Lameter
2007-05-31  7:11               ` Gleb Natapov
2007-05-31  7:24                 ` Christoph Lameter
2007-05-31  7:39                   ` Gleb Natapov
2007-05-31 17:43                     ` Christoph Lameter
2007-05-31 17:07                   ` Lee Schermerhorn
2007-05-31 10:43             ` Andi Kleen
2007-05-31 11:04               ` Gleb Natapov [this message]
2007-05-31 11:30                 ` Gleb Natapov
2007-05-31 15:26                   ` Lee Schermerhorn
2007-05-31 17:41                     ` Gleb Natapov
2007-05-31 18:56                       ` Lee Schermerhorn
2007-05-31 20:06                         ` Gleb Natapov
2007-05-31 20:43                           ` Andi Kleen
2007-06-01  9:38                             ` Gleb Natapov
2007-06-01 10:21                               ` Andi Kleen
2007-06-01 12:25                                 ` Gleb Natapov
2007-06-01 13:09                                   ` Andi Kleen
2007-06-01 17:15                                 ` Lee Schermerhorn
2007-06-01 18:43                                   ` Christoph Lameter
2007-06-01 19:38                                     ` Lee Schermerhorn
2007-06-01 19:48                                       ` Christoph Lameter
2007-06-01 21:05                                         ` Lee Schermerhorn
2007-06-01 21:56                                           ` Christoph Lameter
2007-06-04 13:46                                             ` Lee Schermerhorn
2007-06-04 16:34                                               ` Christoph Lameter
2007-06-04 17:02                                                 ` Lee Schermerhorn
2007-06-04 17:11                                                   ` Christoph Lameter
2007-06-04 20:23                                                     ` Andi Kleen
2007-06-04 21:51                                                       ` Christoph Lameter
2007-06-05 14:30                                                         ` Lee Schermerhorn
2007-06-01 20:28                                     ` Gleb Natapov
2007-06-01 20:45                                       ` Christoph Lameter
2007-06-01 21:10                                         ` Lee Schermerhorn
2007-06-01 21:58                                           ` Christoph Lameter
2007-06-02  7:23                                         ` Gleb Natapov
2007-05-31 11:47                 ` Andi Kleen
2007-05-31 11:59                   ` Gleb Natapov
2007-05-31 12:15                     ` Andi Kleen
2007-05-31 12:18                       ` Gleb Natapov
2007-05-31 18:28       ` Lee Schermerhorn
2007-05-31 18:35         ` Christoph Lameter
2007-05-31 19:29           ` Lee Schermerhorn
2007-05-31 19:25       ` Paul Jackson
2007-05-31 20:22         ` Lee Schermerhorn
2007-05-29 20:07 ` Andi Kleen
2007-05-30 16:04   ` Lee Schermerhorn

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=20070531110412.GM4715@minantech.com \
    --to=glebn@voltaire.com \
    --cc=Lee.Schermerhorn@hp.com \
    --cc=ak@suse.de \
    --cc=akpm@linux-foundation.org \
    --cc=clameter@sgi.com \
    --cc=linux-mm@kvack.org \
    /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