Re: [RFC] extents support for EXT3

From: Ed Sweetman (ed.sweetman_at_wmich.edu)
Date: 08/28/03

  • Next message: Mikael Pettersson: "Re: [PATCH] bad definition of cpus_complement" 
    Date:	Thu, 28 Aug 2003 14:12:21 -0400
To: Alex Tomas <bzzz@tmi.comex.ru>

    This patch seems to be against test2, just wondering if anyone before me
    has used it on the latest test release. If not then i'm gonna do
    something stupid and be the first.

    Alex Tomas wrote:
    > this is 2nd version of the patch. changes:
    > - error handling seems completed
    > - lots of cleanups and comments
    > - few minor bug fixed
    >
    > this version of the patch tries to solve couple
    > of corner cases:
    > - very long truncate
    > - rewrite
    >
    > it survived dbench, bonnie++ and fsx tests.
    >
    > take a look at numbers I've just got, please.
    >
    > before after
    > 5GB file, creation: 2m31.197s 2m21.933s
    > 5GB file, read: 2m25.439s 2m24.833s
    > 5GB file, rewrite: 2m48.434s 2m20.958s
    > 5GB file, removal: 0m8.760s 0m0.858s
    >
    > before after
    > dbench 16: 99.9868 MB/sec 179.243 MB/sec 16 procs
    > dbench 16: 89.9919 MB/sec 203.119 MB/sec 16 procs
    > dbench 16: 73.5519 MB/sec 185.815 MB/sec 16 procs
    > dbench 16: 94.6312 MB/sec 188.519 MB/sec 16 procs
    >
    >
    > to use extents you have to use 'extents' mount option
    >
    > Alex

    -
    To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
    the body of a message to majordomo@vger.kernel.org
    More majordomo info at http://vger.kernel.org/majordomo-info.html
    Please read the FAQ at http://www.tux.org/lkml/

  • Next message: Mikael Pettersson: "Re: [PATCH] bad definition of cpus_complement"

    Relevant Pages

    • [PATCH 5/6][TAKE4] ext4: fallocate support in ext4
      ... This patch implements ->fallocateinode operation in ext4. ... Current implementation only supports preallocation for regular files ... Converted macros that handle uninitialized extents into inline ... if (block + num < end) ...
      (Linux-Kernel)
    • [PATCH 4/5][TAKE3] ext4: fallocate support in ext4
      ... This patch implements ->fallocateinode operation in ext4. ... Current implementation only supports preallocation for regular files ... Converted macros that handle uninitialized extents into inline ... if (block + num < end) ...
      (Linux-Kernel)
    • Re: [PATCH] EXT3 extents against 2.6.0-test7
      ... >>none of my directories have more than 60 or so entries. ... i formatted those partitions with a 4KB inode size. ... extents crashed during readdirsyscall. ... this code patch is very very simple and quite good tested. ...
      (Linux-Kernel)
    • Re: [Ext2-devel] Re: [RFC] extents support for EXT3
      ... > I would not recommend using the patch for system directories only ... this patch seemingly breaks that. ... > Because of that it doesn't seem to be ext3 anymore, ... So what this will be is ext2 format with extents, ...
      (Linux-Kernel)
    • Re: [Suspend2][ 0/9] Extents support.
      ... iterations to break them into manageable chunks. ... Managable chunks means logical changes go together, ... logical change, so it's one patch. ... I must admit I thought you were kidding when I read through this extents ...
      (Linux-Kernel)