linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Fengguang Wu <fengguang.wu@gmail.com>
To: Christoph Lameter <clameter@sgi.com>
Cc: Mel Gorman <mel@skynet.ie>,
	linux-mm@kvack.org, William Lee Irwin III <wli@holomorphy.com>,
	Badari Pulavarty <pbadari@gmail.com>, David Chinner <dgc@sgi.com>,
	Jens Axboe <jens.axboe@oracle.com>,
	Adam Litke <aglitke@gmail.com>, Dave Hansen <hansendc@us.ibm.com>,
	Avi Kivity <avi@argo.co.il>
Subject: Re: [RFC 10/16] Variable Order Page Cache: Readahead fixups
Date: Tue, 22 May 2007 08:59:03 +0800	[thread overview]
Message-ID: <379795533.26423@ustc.edu.cn> (raw)
Message-ID: <20070522005903.GA6184@mail.ustc.edu.cn> (raw)
In-Reply-To: <Pine.LNX.4.64.0705210947450.25871@schroedinger.engr.sgi.com>

On Mon, May 21, 2007 at 09:53:18AM -0700, Christoph Lameter wrote:
> On Mon, 21 May 2007, Fengguang Wu wrote:
> 
> > > I am not sure how to solve that one yet. With the above fix we stay at the 
> > > 2M sized readahead. As the compound order increases so the number of pages
> > > is reduced. We could keep the number of pages constant but then very high
> > > orders may cause a excessive use of memory for readahead.
> > 
> > Do we need to support very high orders(i.e. >2MB)?
> 
> Yes actually we could potentially be using up to 1 TB page size on our 
> new machines that can support several petabytes of RAM. But the read 
> ahead is likely irrelevant in that case. And this is an extreme case that 
> will be rarely used but a customer has required that we will be able to 
> handle such a situation. I think 2-4 megabytes may be more typical.

hehe, 1TB page size is amazing.

> > If not, we can define a MAX_PAGE_CACHE_SIZE=2MB, and limit page orders
> > under that threshold. Now large readahead can be done in
> > MAX_PAGE_CACHE_SIZE chunks.
> 
> Maybe we can just logarithmically decrease the pages for readahead? 
> Readahead should possibly depend on the overall memory of the machine. If 
> the machine has several terabytes of main memory then a couple megs of 
> readahead may be necessary.

Readahead size can be easily scale down by:

 void file_ra_state_init(struct file_ra_state *ra, struct address_space *mapping)
 {
-       ra->ra_pages = mapping->backing_dev_info->ra_pages;
+       ra->ra_pages = DIV_ROUND_UP(mapping->backing_dev_info->ra_pages,
+                                   page_cache_size(mapping));
        ra->prev_index = -1;
 }


But it's not about simply decreasing/disabling readahead.

The problem is, we at least bring in one page at a time.
It's not a problem for 2-4MB page sizes.
But to support page size up to 1TB, this behavior must be changed.

--
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:[~2007-05-22  0:59 UTC|newest]

Thread overview: 40+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-04-23  6:48 [RFC 00/16] Variable Order Page Cache Patchset V2 Christoph Lameter
2007-04-23  6:48 ` [RFC 01/16] Free up page->private for compound pages Christoph Lameter
2007-04-24  2:12   ` Dave Hansen
2007-04-24  2:23     ` Christoph Lameter
2007-04-25 10:55   ` Mel Gorman
2007-04-23  6:48 ` [RFC 02/16] vmstat.c: Support accounting " Christoph Lameter
2007-04-25 10:59   ` Mel Gorman
2007-04-25 15:43     ` Christoph Lameter
2007-04-23  6:49 ` [RFC 03/16] Variable Order Page Cache: Add order field in mapping Christoph Lameter
2007-04-25 11:05   ` Mel Gorman
2007-04-23  6:49 ` [RFC 04/16] Variable Order Page Cache: Add basic allocation functions Christoph Lameter
2007-04-23  6:49 ` [RFC 05/16] Variable Order Page Cache: Add functions to establish sizes Christoph Lameter
2007-04-25 11:20   ` Mel Gorman
2007-04-25 15:54     ` Christoph Lameter
2007-04-23  6:49 ` [RFC 06/16] Variable Page Cache: Add VM_BUG_ONs to check for correct page order Christoph Lameter
2007-04-25 11:22   ` Mel Gorman
2007-04-23  6:49 ` [RFC 07/16] Variable Order Page Cache: Add clearing and flushing function Christoph Lameter
2007-04-23  6:49 ` [RFC 08/16] Variable Order Page Cache: Fixup fallback functions Christoph Lameter
2007-04-23  6:49 ` [RFC 09/16] Variable Order Page Cache: Fix up mm/filemap.c Christoph Lameter
2007-04-23  6:49 ` [RFC 10/16] Variable Order Page Cache: Readahead fixups Christoph Lameter
2007-04-25 11:36   ` Mel Gorman
2007-04-25 15:56     ` Christoph Lameter
     [not found]       ` <20070521104204.GA8795@mail.ustc.edu.cn>
2007-05-21 10:42         ` Fengguang Wu
2007-05-21 16:53           ` Christoph Lameter
     [not found]             ` <20070522005903.GA6184@mail.ustc.edu.cn>
2007-05-22  0:59               ` Fengguang Wu [this message]
     [not found]             ` <20070524040453.GA10662@mail.ustc.edu.cn>
2007-05-24  4:04               ` Fengguang Wu
2007-05-24  4:06                 ` Christoph Lameter
2007-04-23  6:49 ` [RFC 11/16] Variable Page Cache Size: Fix up reclaim counters Christoph Lameter
2007-04-25 13:08   ` Mel Gorman
2007-04-23  6:49 ` [RFC 12/16] Variable Order Page Cache: Fix up the writeback logic Christoph Lameter
2007-04-23  6:49 ` [RFC 13/16] Variable Order Page Cache: Fixed to block layer Christoph Lameter
2007-04-23  6:49 ` [RFC 14/16] Variable Order Page Cache: Add support to ramfs Christoph Lameter
2007-04-23  6:50 ` [RFC 15/16] ext2: Add variable page size support Christoph Lameter
2007-04-23 16:30   ` Badari Pulavarty
2007-04-24  1:11     ` Christoph Lameter
2007-04-23  6:50 ` [RFC 16/16] Variable Order Page Cache: Alternate implementation of page cache macros Christoph Lameter
2007-04-25 13:16   ` Mel Gorman
2007-04-23  9:23 ` [RFC 00/16] Variable Order Page Cache Patchset V2 David Chinner
2007-04-23  9:31 ` David Chinner
  -- strict thread matches above, loose matches on Subject: below --
2007-04-23  6:21 clameter
2007-04-23  6:21 ` [RFC 10/16] Variable Order Page Cache: Readahead fixups clameter

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=379795533.26423@ustc.edu.cn \
    --to=fengguang.wu@gmail.com \
    --cc=aglitke@gmail.com \
    --cc=avi@argo.co.il \
    --cc=clameter@sgi.com \
    --cc=dgc@sgi.com \
    --cc=hansendc@us.ibm.com \
    --cc=jens.axboe@oracle.com \
    --cc=linux-mm@kvack.org \
    --cc=mel@skynet.ie \
    --cc=pbadari@gmail.com \
    --cc=wli@holomorphy.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox