linux-ext4.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Alex Tomas <bzzz.tomas@gmail.com>
To: Eric Sandeen <sandeen@redhat.com>
Cc: Andreas Dilger <adilger@Sun.COM>,
	ext4 development <linux-ext4@vger.kernel.org>
Subject: Re: delalloc fragmenting files?
Date: Thu, 01 Nov 2007 03:35:46 +0300	[thread overview]
Message-ID: <47291F62.3000008@gmail.com> (raw)
In-Reply-To: <4728ED3B.1060407@redhat.com>

[-- Attachment #1: Type: text/plain, Size: 600 bytes --]

please, try the patch attached.

thanks, Alex

Eric Sandeen wrote:
> One thing that seems to be happening is that thanks to delalloc, a nice
> big request is coming in (only 1036 blocks of the 4096, not quite sure
> why), but then it gets into ext4_mb_normalize_request(), which finds the
> most blocks it can "preallocate" is 256, and chops down the request to
> 256 blocks.  Shouldn't this preallocation be over & above what was asked
> for, vs. reducing the request?
> 
> Ok, so, we get allocations in 256-block chunks...  Why they don't all
> come out contiguous, I don't know yet...
> 
> -Eric


[-- Attachment #2: mballoc-debug.patch --]
[-- Type: text/x-patch, Size: 901 bytes --]

Index: linux-2.6.24-rc1/fs/ext4/mballoc.c
===================================================================
--- linux-2.6.24-rc1.orig/fs/ext4/mballoc.c	2007-10-27 10:29:17.000000000 +0400
+++ linux-2.6.24-rc1/fs/ext4/mballoc.c	2007-10-27 22:14:54.000000000 +0400
@@ -3088,8 +3088,10 @@ static void ext4_mb_normalize_request(st
 			break;
 		}
 	}
+	size = wind;
+
 	if (wind == 0) {
-		__u64 tstart;
+		__u64 tstart, tend;
 		/* file is quite large, we now preallocate with
 		 * the biggest configured window with regart to
 		 * logical offset */
@@ -3097,8 +3099,11 @@ static void ext4_mb_normalize_request(st
 		tstart = ac->ac_o_ex.fe_logical;
 		do_div(tstart, wind);
 		start = tstart * wind;
+		tend = ac->ac_o_ex.fe_logical + ac->ac_o_ex.fe_len - 1;
+		do_div(tend, wind);
+		tend = tend * wind + wind;
+		size = tend - start;
 	}
-	size = wind;
 	orig_size = size;
 	orig_start = start;
 

  reply	other threads:[~2007-11-01  0:36 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-10-26 21:24 delalloc fragmenting files? Eric Sandeen
2007-10-26 22:19 ` Andreas Dilger
2007-10-27  2:54   ` Eric Sandeen
2007-10-27  3:08     ` Eric Sandeen
2007-10-31 21:01   ` Eric Sandeen
2007-11-01  0:35     ` Alex Tomas [this message]
2007-11-01 15:20       ` Eric Sandeen
2007-11-01 19:05         ` Eric Sandeen
2007-11-02  8:56           ` Alex Tomas
2007-11-02 15:41             ` Eric Sandeen
2007-11-04 18:36               ` Alex Tomas
2007-11-05 17:37                 ` Eric Sandeen
2007-11-05 17:39                   ` Eric Sandeen
2007-11-05 17:45                   ` Alex Tomas
2007-11-05 18:51                   ` Andreas Dilger
2007-11-05 18:54                     ` Eric Sandeen
2007-11-06 19:54                   ` Eric Sandeen
2007-11-07 20:37                     ` Andreas Dilger
2007-11-07 20:53                       ` Eric Sandeen
2007-10-29 19:13 ` Eric Sandeen

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=47291F62.3000008@gmail.com \
    --to=bzzz.tomas@gmail.com \
    --cc=adilger@Sun.COM \
    --cc=linux-ext4@vger.kernel.org \
    --cc=sandeen@redhat.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;
as well as URLs for NNTP newsgroup(s).