All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Martin K. Petersen" <martin.petersen@oracle.com>
To: Mark Lord <kernel@teksavvy.com>
Cc: "Ted Ts'o" <tytso@mit.edu>, Lukas Czerner <lczerner@redhat.com>,
	Steven Whitehouse <swhiteho@redhat.com>,
	James Bottomley <James.Bottomley@suse.de>,
	Christoph Hellwig <hch@infradead.org>,
	Matthew Wilcox <matthew@wil.cx>, Josef Bacik <josef@redhat.com>,
	linux-ext4@vger.kernel.org, linux-kernel@vger.kernel.org,
	linux-fsdevel@vger.kernel.org, sandeen@redhat.com
Subject: Re: [PATCH 1/2] fs: Do not dispatch FITRIM through separate super_operation
Date: Wed, 24 Nov 2010 23:23:34 -0500	[thread overview]
Message-ID: <yq18w0ini9l.fsf@sermon.lab.mkp.net> (raw)
In-Reply-To: <4CEDCE60.9050109@teksavvy.com> (Mark Lord's message of "Wed, 24 Nov 2010 21:48:00 -0500")

>>>>> "Mark" == Mark Lord <kernel@teksavvy.com> writes:

Mark> But mke2fs probably is NOT doing a "single big discard", because
Mark> for SATA the TRIM command is limited to 64K sectors per
Mark> range.. and the in-kernel TRIM code only ever does single ranges..

How many times must I reiterate that this is not true?

Due to various limitations in the way the block layer works we can not
clear LBA 10 to 20 and LBA 40 to 50 in one command. We don't have an
infrastructure in place that allows us to operate on several discrete
areas in one request.

But for contiguous areas we will cram as many ranges entries in as we
can fit in the TRIM payload. This means we'll issue one TRIM command for
every 2 GiB minus change (65535*64*512 / 1048576 = 2047 MiB).

In your example above blkdev_issue_discard() will loop over the 100 GiB
range requested by the caller (filesystem or BLKDISCARD ioctl) and issue
about fifty TRIM commands to the drive.

Large discards are a best-case scenario (short of SECURE ERASE). So if
your drive takes forever to perform this operation then all bets are
off in the TRIM department.


Mark> So doing a discard over an entire drive-encompassing partition,
Mark> say.. 100GB, will require 3000+ individual TRIM commands. 

No.

-- 
Martin K. Petersen	Oracle Linux Engineering

  reply	other threads:[~2010-11-25  4:24 UTC|newest]

Thread overview: 106+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-11-18  7:36 [PATCH 1/2] fs: Do not dispatch FITRIM through separate super_operation Lukas Czerner
2010-11-18  7:36 ` [PATCH 2/2] ext4: Add EXT4_IOC_TRIM ioctl to handle batched discard Lukas Czerner
2010-11-19 16:19   ` Ted Ts'o
2010-11-19 16:26     ` Lukas Czerner
2010-11-20  1:37       ` Ted Ts'o
2010-11-18 13:06 ` [PATCH 1/2] fs: Do not dispatch FITRIM through separate super_operation Matthew Wilcox
2010-11-18 13:48   ` Josef Bacik
2010-11-18 14:19     ` Matthew Wilcox
2010-11-18 14:29       ` Christoph Hellwig
2010-11-18 17:19         ` James Bottomley
2010-11-18 17:22           ` Jeff Moyer
2010-11-18 17:41             ` James Bottomley
2010-11-18 20:04               ` Greg Freemyer
2010-11-18 20:04                 ` Greg Freemyer
2010-11-18 21:42                 ` Mark Lord
2010-11-18 21:44                 ` Mark Lord
2010-11-18 21:50                   ` James Bottomley
2010-11-18 22:07                     ` Mark Lord
2010-11-19  1:33                       ` Ted Ts'o
2010-11-19  3:44                         ` Mark Lord
2010-11-19  3:44                         ` Mark Lord
2010-11-19  3:44                           ` Mark Lord
2010-11-19 13:58                           ` Jeff Moyer
2010-11-18 23:52                   ` Martin K. Petersen
2010-11-19  0:34                     ` Mark Lord
2010-11-19  1:16                       ` Greg Freemyer
2010-11-19  1:16                         ` Greg Freemyer
2010-11-19 11:55                         ` Christoph Hellwig
2010-11-19 14:01                           ` Mark Lord
2010-11-19 14:06                             ` Christoph Hellwig
2010-11-19 14:48                               ` Mark Lord
2010-11-19 14:54                                 ` Christoph Hellwig
2010-11-19 15:24                                   ` Mark Lord
2010-11-19 15:34                                     ` Christoph Hellwig
2010-11-19 16:20                           ` Greg Freemyer
2010-11-19 16:38                             ` Christoph Hellwig
2010-11-19 18:06                               ` Lukas Czerner
2010-11-19 18:10                                 ` Lukas Czerner
2010-11-19 18:14                                   ` Lukas Czerner
2010-11-19 19:29                                 ` Chris Mason
2010-11-19  1:49                       ` Martin K. Petersen
2010-11-19  3:42                         ` Mark Lord
2010-11-18 18:05             ` Jamie Lokier
2010-11-18 19:32               ` Markus Trippelsdorf
2010-11-18 21:45                 ` Mark Lord
2010-11-18 21:50                   ` Markus Trippelsdorf
2010-11-18 22:09                     ` Mark Lord
2010-11-18 17:35           ` Lukas Czerner
2010-11-19 12:16             ` Steven Whitehouse
2010-11-19 13:53               ` Mark Lord
2010-11-19 14:02                 ` Ted Ts'o
2010-11-19 14:10                   ` Christoph Hellwig
2010-11-19 14:10                   ` Christoph Hellwig
2010-11-19 14:10                   ` Christoph Hellwig
2010-11-19 15:37                     ` Ted Ts'o
2010-11-19 15:50                       ` Christoph Hellwig
2010-11-19 15:50                       ` Christoph Hellwig
2010-11-19 15:50                         ` Christoph Hellwig
2010-11-19 16:16                         ` Ted Ts'o
2010-11-19 14:50                   ` Mark Lord
2010-11-19 14:50                   ` Mark Lord
2010-11-19 14:50                     ` Mark Lord
2010-11-19 15:35                   ` Mark Lord
2010-11-19 15:35                   ` Mark Lord
2010-11-19 15:35                     ` Mark Lord
2010-11-19 15:44                     ` Lukas Czerner
2010-11-19 16:30                       ` Ted Ts'o
2010-11-19 22:49                         ` Mark Lord
2010-11-19 22:49                           ` Mark Lord
2010-11-19 22:49                         ` Mark Lord
2010-11-25  2:48                         ` Mark Lord
2010-11-25  2:48                         ` Mark Lord
2010-11-25  2:48                           ` Mark Lord
2010-11-25  4:23                           ` Martin K. Petersen [this message]
2010-11-25 14:44                             ` Mark Lord
2010-11-25  4:41                           ` Greg Freemyer
2010-11-25  4:41                             ` Greg Freemyer
2010-11-25 14:53                             ` Mark Lord
2010-11-25 16:24                               ` Greg Freemyer
2010-11-25 16:24                                 ` Greg Freemyer
2010-11-26 13:49                                 ` Mark Lord
2010-11-26 14:00                                   ` Lukas Czerner
2010-11-18 17:55           ` Chris Mason
2010-12-03 18:24             ` Ric Wheeler
2010-11-18 21:37           ` Mark Lord
2010-11-19 11:09             ` Christoph Hellwig
2010-11-19 13:54               ` Mark Lord
2010-11-19 14:40             ` Chris Mason
2010-11-19 14:53               ` Mark Lord
2010-11-19 14:57                 ` Christoph Hellwig
2010-11-19 15:21                   ` Mark Lord
2010-12-07  9:27                     ` Christoph Hellwig
2010-12-07 16:52                       ` Chris Mason
2011-06-02  4:52                         ` Kyungmin Park
2011-06-02  8:14                           ` Lukas Czerner
2011-06-03  2:06                           ` Dave Chinner
2011-06-03  2:06                             ` Dave Chinner
2011-06-03  4:25                             ` Kyungmin Park
2011-06-03  4:25                               ` Kyungmin Park
2010-11-19 15:30                   ` Mark Lord
2010-11-21 19:07                 ` Valdis.Kletnieks
2010-11-21 20:20                   ` James Bottomley
2010-11-18 14:31       ` Josef Bacik
2010-11-18 14:36         ` Tao Ma
2010-11-19 15:41 ` Ted Ts'o
2010-11-19 15:50   ` Christoph Hellwig

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=yq18w0ini9l.fsf@sermon.lab.mkp.net \
    --to=martin.petersen@oracle.com \
    --cc=James.Bottomley@suse.de \
    --cc=hch@infradead.org \
    --cc=josef@redhat.com \
    --cc=kernel@teksavvy.com \
    --cc=lczerner@redhat.com \
    --cc=linux-ext4@vger.kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=matthew@wil.cx \
    --cc=sandeen@redhat.com \
    --cc=swhiteho@redhat.com \
    --cc=tytso@mit.edu \
    /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.