public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Andrew Morton <akpm@osdl.org>
To: markw@osdl.org
Cc: linux-kernel@vger.kernel.org
Subject: Re: 2.6.4-mm2
Date: Thu, 18 Mar 2004 10:06:15 -0800	[thread overview]
Message-ID: <20040318100615.7f2943ea.akpm@osdl.org> (raw)
In-Reply-To: <200403181737.i2IHbCE09261@mail.osdl.org>

markw@osdl.org wrote:
>
> Sorry I'm falling behind...  I see about a 10% decrease in throughput
> with our dbt2 workload when comparing 2.6.4-mm2 to 2.6.3.  I'm wondering
> if this might be a result of the changes to the pagecache, radix-tree
> and writeback code since you mentioned it could affect i/o scheduling in
> 2.6.4-mm1.

Could be.  Have you run tests without LVM in the picture?

> PostgreSQL is using 8KB blocks and the characteristics of the i/o should
> be that one lvm2 volume is experiencing mostly sequential writes, while
> another has random reading and writing.  Both these volumes are using
> ext2.  I'll summarize the throughput results here, with the lvm2 stripe
> width varying across the columns:
> 
> kernel          16 kb   32 kb   64 kb   128 kb  256 kb  512 kb
> 2.6.3                           2308    2335    2348    2334
> 2.6.4-mm2       2028    2048    2074    2096    2082    2078
> 
> Here's a page with links to profile, oprofile, etc of each result:
> 	http://developer.osdl.org/markw/linux/2.6-pagecache.html
> 
> Comparing one pair of readprofile results, I find it curious that
> dm_table_unplug_all and dm_table_any_congested show up near the top of a
> 2.6.4-mm2 profile when they haven't shown up before in 2.6.3.

14015190 poll_idle                                241641.2069
175162 generic_unplug_device                    1317.0075
165480 __copy_from_user_ll                      1272.9231
161151 __copy_to_user_ll                        1342.9250
152106 schedule                                  85.0705
142395 DAC960_LP_InterruptHandler               761.4706
113677 dm_table_unplug_all                      1386.3049
 65420 __make_request                            45.5571
 64832 dm_table_any_congested                   697.1183
 37913 try_to_wake_up                            32.2939

That's broken.  How many disks are involve in the DM stack?

The relevant code was reworked subsequent to 2.6.4-mm2.  Maybe we fixed
this, but I cannot immediately explain what you're seeing here.


  reply	other threads:[~2004-03-18 18:06 UTC|newest]

Thread overview: 90+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2004-03-15  1:28 2.6.4-mm2 Andrew Morton
2004-03-15  1:57 ` 2.6.4-mm2 Joshua Kwan
2004-03-15  6:08   ` 2.6.4-mm2 Sam Ravnborg
2004-03-15  9:32 ` [patch] 2.6.4-mm2: ALSA au88x0.c doesn't compile with gcc 2.95 Adrian Bunk
2004-03-15  9:36 ` 2.6.4-mm2: ALSA au88{1,2}0: multiply defined symbols Adrian Bunk
2004-03-15 18:54 ` 2.6.4-mm2 Sam Ravnborg
2004-03-20 22:50   ` 2.6.4-mm2 Olaf Hering
2004-03-15 19:18 ` 2.6.4-mm2 (compile stats) John Cherry
2004-03-15 20:57 ` 2.6.4-mm2 Thomas Schlichter
2004-03-15 21:08   ` 2.6.4-mm2 Thomas Schlichter
2004-03-15 21:18   ` 2.6.4-mm2 Andrew Morton
2004-03-15 21:18     ` 2.6.4-mm2 Jens Axboe
2004-03-15 22:40     ` 2.6.4-mm2 Thomas Schlichter
2004-03-15 22:11 ` 2.6.4-mm2 Jesse Barnes
2004-03-16 18:32 ` 2.6.4-mm2 Daniel McNeil
2004-03-16 21:58   ` 2.6.4-mm2 Chris Mason
2004-03-16 23:21     ` 2.6.4-mm2 Andrew Morton
2004-03-16 23:28       ` 2.6.4-mm2 Andrew Morton
2004-03-16 23:39         ` 2.6.4-mm2 Andrew Morton
2004-03-17  0:10           ` 2.6.4-mm2 Daniel McNeil
     [not found]           ` <1079485055.4181.1115.camel@watt.suse.com>
     [not found]             ` <1079487710.3100.22.camel@ibm-c.pdx.osdl.net>
     [not found]               ` <20040316180043.441e8150.akpm@osdl.org>
2004-03-17 20:11                 ` 2.6.4-mm2 Chris Mason
2004-03-17 20:33                   ` 2.6.4-mm2 Andrew Morton
2004-03-17 22:46                     ` 2.6.4-mm2 Chris Mason
2004-03-17 23:09                       ` 2.6.4-mm2 Andrew Morton
2004-03-17 23:27                         ` 2.6.4-mm2 Chris Mason
2004-03-17 23:51                           ` 2.6.4-mm2 Andrew Morton
2004-03-18  0:06                             ` 2.6.4-mm2 Chris Mason
2004-03-18  0:13                               ` 2.6.4-mm2 Andrew Morton
2004-03-18  0:31                                 ` 2.6.4-mm2 Chris Mason
2004-03-18  0:33                                   ` 2.6.4-mm2 Andrew Morton
2004-03-18  0:41                                     ` 2.6.4-mm2 Chris Mason
2004-03-18  1:15                                     ` 2.6.4-mm2 Daniel McNeil
2004-03-18 17:53                                       ` 2.6.4-mm2 Daniel McNeil
2004-03-18 18:47                                         ` 2.6.4-mm2 Chris Mason
2004-03-18 19:10                                           ` 2.6.4-mm2 Daniel McNeil
2004-03-19 16:49                                             ` 2.6.5-rc2-mm2 and direct_read_under Daniel McNeil
2004-03-19 17:05                                               ` 2.6.5-rc1-mm2 " Daniel McNeil
2004-03-21 14:36                                                 ` Chris Mason
2004-03-22 18:10                                                   ` 2.6.5-rc1-mm2 and direct_read_under and wb Daniel McNeil
2004-03-22 18:23                                                     ` Chris Mason
2004-03-22 18:27                                                       ` Chris Mason
2004-03-22 18:35                                                     ` Chris Mason
2004-03-22 18:51                                                       ` Daniel McNeil
2004-03-22 23:13                                                         ` Andrew Morton
2004-03-23  0:51                                                           ` Daniel McNeil
2004-03-23  9:25                                                             ` Andrew Morton
2004-03-23 17:05                                                               ` Daniel McNeil
2004-03-23 17:59                                                                 ` Andrew Morton
2004-03-23 21:38                                                                   ` Daniel McNeil
2004-03-23 21:47                                                                     ` Andrew Morton
2004-03-18 17:37 ` 2.6.4-mm2 markw
2004-03-18 18:06   ` Andrew Morton [this message]
2004-03-18 18:48     ` 2.6.4-mm2 markw
2004-03-18 19:10       ` 2.6.4-mm2 Chris Mason
2004-03-18 19:27     ` 2.6.4-mm2 Jens Axboe
2004-03-18 23:38       ` 2.6.4-mm2 markw
2004-03-19  7:39         ` 2.6.4-mm2 Jens Axboe
2004-03-19  3:15       ` 2.6.4-mm2 Andrew Morton
2004-03-19  7:39         ` 2.6.4-mm2 Jens Axboe
2004-03-19  7:52           ` 2.6.4-mm2 Andrew Morton
2004-03-19  7:57             ` 2.6.4-mm2 Jens Axboe
2004-03-19  8:19               ` 2.6.4-mm2 Andrew Morton
2004-03-19  8:31                 ` 2.6.4-mm2 Jens Axboe
2004-03-19  8:39                   ` 2.6.4-mm2 Andrew Morton
2004-03-19  8:48                     ` 2.6.4-mm2 Jens Axboe
2004-03-19  9:56             ` 2.6.4-mm2 Miquel van Smoorenburg
2004-03-19 10:00               ` 2.6.4-mm2 Jens Axboe
     [not found]         ` <20040318194150.4de65049.akpm@osdl.org>
2004-03-20  2:39           ` 2.6.4-mm2 Mark Wong
2004-03-20  2:47             ` 2.6.4-mm2 Mark Wong
2004-03-20  2:50             ` 2.6.4-mm2 Andrew Morton
2004-03-20  2:53               ` 2.6.4-mm2 Mark Wong
2004-03-20  3:52                 ` 2.6.4-mm2 Nick Piggin
2004-03-20  4:14                   ` 2.6.4-mm2 Andrew Morton
2004-03-20  4:24                     ` 2.6.4-mm2 Nick Piggin
2004-03-20  4:26                       ` 2.6.4-mm2 Nick Piggin
2004-03-20 21:17                       ` 2.6.4-mm2 Martin J. Bligh
2004-03-22 17:19             ` 2.6.4-mm2 Mary Edie Meredith
2004-03-23  0:27               ` 2.6.4-mm2 Andrew Morton
2004-03-23 19:21                 ` 2.6.4-mm2 Mary Edie Meredith
2004-03-23 19:32                   ` 2.6.4-mm2 Andrew Morton
2004-03-24  0:07                     ` 2.6.4-mm2 Mary Edie Meredith
2004-03-30 21:30                       ` 2.6.4-mm2 Mary Edie Meredith
     [not found] <A6974D8E5F98D511BB910002A50A6647615F5E26@hdsmsx402.hd.intel.com>
2004-03-20  4:19 ` 2.6.4-mm2 Len Brown
2004-03-20  4:26   ` 2.6.4-mm2 Andrew Morton
2004-03-20  4:32     ` 2.6.4-mm2 Mark Wong
     [not found] <A6974D8E5F98D511BB910002A50A6647615F5E2B@hdsmsx402.hd.intel.com>
2004-03-20  4:27 ` 2.6.4-mm2 Len Brown
2004-03-20  9:01   ` 2.6.4-mm2 Nick Piggin
2004-03-22 16:24     ` 2.6.4-mm2 markw
  -- strict thread matches above, loose matches on Subject: below --
2004-03-20 23:12 2.6.4-mm2 sam
2004-03-20 23:41 ` 2.6.4-mm2 Olaf Hering

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=20040318100615.7f2943ea.akpm@osdl.org \
    --to=akpm@osdl.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=markw@osdl.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