From: "Lukáš Czerner" <lczerner@redhat.com>
To: "Darrick J. Wong" <darrick.wong@oracle.com>
Cc: tytso@mit.edu, linux-ext4@vger.kernel.org
Subject: Re: [PATCH 15/37] mke2fs: set block_validity as a default mount option
Date: Mon, 5 May 2014 19:24:14 +0200 (CEST) [thread overview]
Message-ID: <alpine.LFD.2.00.1405051921290.31228@localhost.localdomain> (raw)
In-Reply-To: <20140501231400.31890.27469.stgit@birch.djwong.org>
On Thu, 1 May 2014, Darrick J. Wong wrote:
> Date: Thu, 01 May 2014 16:14:00 -0700
> From: Darrick J. Wong <darrick.wong@oracle.com>
> To: tytso@mit.edu, darrick.wong@oracle.com
> Cc: linux-ext4@vger.kernel.org
> Subject: [PATCH 15/37] mke2fs: set block_validity as a default mount option
>
> The block_validity mount option spot-checks block allocations against
> a bitmap of known group metadata blocks. This helps us to prevent
> self-inflicted catastrophic failures such as trying to "share"
> critical metadata (think bitmaps) with file data, which usually
> results in filesystem destruction.
>
> In order to test the overhead of the mount option, I re-used the speed
> tests in the metadata checksum testing script. In short, the program
> creates what looks like 15 copies of a kernel source tree, except that
> it uses fallocate to strip out the overhead of writing the file data
> so that we can focus on metadata overhead. On a 64G RAM disk, the
> overhead was generally about 0.9% and at most 1.6%. On a 160G USB
> disk, the overhead was about 0.8% and peaked at 1.2%.
>
> When I changed the test to write out files instead of merely
> fallocating space, the overhead was negligible.
I like that, but I think we need to run more performance testing on
that to make sure that it really does not break something.
Eric, will you be able to run your performance tests with
block_validity option and compare it with baseline to see whether it
really does not change anything ?
I'll try to run my tests as well.
We really need generic performance test suite, I hope that Dave will
send his performance addition to xfstests soon so we can start
adding tests.
Thanks!
-Lukas
>
> Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
> ---
> misc/mke2fs.conf.in | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
>
> diff --git a/misc/mke2fs.conf.in b/misc/mke2fs.conf.in
> index 4c5dba7..de0250d 100644
> --- a/misc/mke2fs.conf.in
> +++ b/misc/mke2fs.conf.in
> @@ -1,6 +1,6 @@
> [defaults]
> base_features = sparse_super,filetype,resize_inode,dir_index,ext_attr
> - default_mntopts = acl,user_xattr
> + default_mntopts = acl,user_xattr,block_validity
> enable_periodic_fsck = 0
> blocksize = 4096
> inode_size = 256
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
next prev parent reply other threads:[~2014-05-05 17:24 UTC|newest]
Thread overview: 91+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-05-01 23:12 [PATCH 00/37] e2fsprogs patchbomb 5/14 Darrick J. Wong
2014-05-01 23:12 ` [PATCH 01/37] misc: create better-packaged static analysis reports Darrick J. Wong
2014-05-11 22:33 ` Theodore Ts'o
2014-05-01 23:12 ` [PATCH 02/37] misc: coverity fixes Darrick J. Wong
2014-05-02 11:17 ` Lukáš Czerner
2014-05-05 20:04 ` Darrick J. Wong
2014-05-11 22:40 ` Theodore Ts'o
2014-05-01 23:12 ` [PATCH 03/37] libext2fs: create sockets when populating filesystem Darrick J. Wong
2014-05-02 11:22 ` Lukáš Czerner
2014-05-05 20:08 ` Darrick J. Wong
2014-05-11 22:44 ` Theodore Ts'o
2014-05-01 23:12 ` [PATCH 04/37] mke2fs: always warn if 128-byte inode and inline_data Darrick J. Wong
2014-05-02 11:27 ` Lukáš Czerner
2014-05-05 20:10 ` Darrick J. Wong
2014-05-12 0:26 ` Theodore Ts'o
2014-05-01 23:12 ` [PATCH 05/37] debugfs: teach logdump to deal with 64bit revoke tables Darrick J. Wong
2014-05-02 11:38 ` Lukáš Czerner
2014-05-05 22:23 ` Darrick J. Wong
2014-05-06 11:35 ` Lukáš Czerner
2014-05-12 1:20 ` Theodore Ts'o
2014-05-01 23:13 ` [PATCH 06/37] debugfs: force logdump to display (old) journal contents Darrick J. Wong
2014-05-02 11:49 ` Lukáš Czerner
2014-05-06 0:24 ` Darrick J. Wong
2014-05-12 1:41 ` Theodore Ts'o
2014-05-12 3:31 ` Theodore Ts'o
2014-05-14 0:05 ` Darrick J. Wong
2014-05-01 23:13 ` [PATCH 07/37] resize2fs: fix check for collision between old GDT and superblock on sparse_super2 fs Darrick J. Wong
2014-05-12 3:35 ` Theodore Ts'o
2014-05-01 23:13 ` [PATCH 08/37] mke2fs: set gdt csum when creating packed fs Darrick J. Wong
2014-05-02 11:55 ` Lukáš Czerner
2014-05-12 4:22 ` Theodore Ts'o
2014-05-01 23:13 ` [PATCH 09/37] mke2fs: set error behavior at initialization time Darrick J. Wong
2014-05-02 12:13 ` Lukáš Czerner
2014-05-01 23:13 ` [PATCH 10/37] e2fsck: verify checksums after checking everything else Darrick J. Wong
2014-05-02 12:32 ` Lukáš Czerner
2014-05-05 22:56 ` Darrick J. Wong
2014-05-06 11:32 ` Lukáš Czerner
2014-05-08 0:05 ` Darrick J. Wong
2014-05-01 23:13 ` [PATCH 11/37] e2fsck: fix the extended attribute checksum error message Darrick J. Wong
2014-05-02 12:46 ` Lukáš Czerner
2014-05-05 23:08 ` Darrick J. Wong
2014-05-06 10:12 ` Lukáš Czerner
2014-05-01 23:13 ` [PATCH 12/37] e2fsck: insert a missing dirent tail for checksums if possible Darrick J. Wong
2014-05-02 12:54 ` Lukáš Czerner
2014-05-05 23:16 ` Darrick J. Wong
2014-05-01 23:13 ` [PATCH 13/37] e2fsck: write dir blocks after new inode when reconstructing root/lost+found Darrick J. Wong
2014-05-05 17:13 ` Lukáš Czerner
2014-05-01 23:13 ` [PATCH 14/37] dumpe2fs: add switch to disable checksum verification Darrick J. Wong
2014-05-05 17:20 ` Lukáš Czerner
2014-05-01 23:14 ` [PATCH 15/37] mke2fs: set block_validity as a default mount option Darrick J. Wong
2014-05-05 17:24 ` Lukáš Czerner [this message]
2014-05-01 23:14 ` [PATCH 16/37] libext2fs: support allocating uninit blocks in bmap2() Darrick J. Wong
2014-05-06 15:45 ` Lukáš Czerner
2014-05-06 19:59 ` Darrick J. Wong
2014-05-07 10:02 ` Lukáš Czerner
2014-05-07 21:37 ` Darrick J. Wong
2014-05-08 0:13 ` [PATCH 1/2] libext2fs: support BLKZEROOUT/FALLOC_FL_ZERO_RANGE in ext2fs_zero_blocks Darrick J. Wong
2014-05-13 11:11 ` Lukáš Czerner
2014-05-08 0:14 ` [PATCH 2/2] libext2fs: support allocating uninit blocks in bmap2() Darrick J. Wong
2014-05-27 16:28 ` Lukáš Czerner
2014-05-28 19:48 ` Darrick J. Wong
2014-05-01 23:14 ` [PATCH 17/37] libext2fs: file IO routines should handle uninit blocks Darrick J. Wong
2014-05-01 23:14 ` [PATCH 18/37] resize2fs: convert fs to and from 64bit mode Darrick J. Wong
2014-05-01 23:14 ` [PATCH 19/37] resize2fs: when toggling 64bit, don't free in-use bg data clusters Darrick J. Wong
2014-05-01 23:14 ` [PATCH 20/37] resize2fs: adjust reserved_gdt_blocks when changing group descriptor size Darrick J. Wong
2014-05-01 23:14 ` [PATCH 21/37] libext2fs: have UNIX IO manager use pread/pwrite Darrick J. Wong
2014-08-02 23:16 ` Theodore Ts'o
2014-05-01 23:14 ` [PATCH 22/37] ext2fs: add readahead method to improve scanning Darrick J. Wong
2014-05-01 23:14 ` [PATCH 23/37] e2fsck: provide routines to read-ahead metadata Darrick J. Wong
2014-05-01 23:14 ` [PATCH 24/37] e2fsck: read-ahead metadata during passes 1, 2, and 4 Darrick J. Wong
2014-07-28 22:25 ` Darrick J. Wong
2014-05-01 23:15 ` [PATCH 25/37] libext2fs: when appending to a file, don't split an index block in equal halves Darrick J. Wong
2014-08-02 23:43 ` Theodore Ts'o
2014-05-01 23:15 ` [PATCH 26/37] libext2fs: find inode goal when allocating blocks Darrick J. Wong
2014-05-01 23:15 ` [PATCH 27/37] libext2fs: find a range of empty blocks Darrick J. Wong
2014-05-01 23:15 ` [PATCH 28/37] libext2fs: provide a function to set inode size Darrick J. Wong
2014-07-26 18:37 ` Theodore Ts'o
2014-05-01 23:15 ` [PATCH 29/37] libext2fs: implement fallocate Darrick J. Wong
2014-05-01 23:15 ` [PATCH 31/37] fuse2fs: translate ACL structures Darrick J. Wong
2014-05-01 23:15 ` [PATCH 32/37] fuse2fs: handle 64-bit dates correctly Darrick J. Wong
2014-05-01 23:16 ` [PATCH 33/37] fuse2fs: implement fallocate Darrick J. Wong
2014-05-01 23:16 ` [PATCH 35/37] tests: enable using fuse2fs with metadata checksum test Darrick J. Wong
2014-05-01 23:16 ` [PATCH 36/37] tests: test date handling Darrick J. Wong
2014-05-01 23:16 ` [PATCH 37/37] ext5: define new subtype to add features and reduce testing complexity Darrick J. Wong
2014-05-02 9:45 ` Lukáš Czerner
2014-05-02 14:04 ` Theodore Ts'o
2014-05-06 1:59 ` Darrick J. Wong
2014-05-06 1:33 ` Darrick J. Wong
2014-05-06 12:50 ` Lukáš Czerner
2014-05-06 15:21 ` Theodore Ts'o
2014-05-06 15:30 ` Lukáš Czerner
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=alpine.LFD.2.00.1405051921290.31228@localhost.localdomain \
--to=lczerner@redhat.com \
--cc=darrick.wong@oracle.com \
--cc=linux-ext4@vger.kernel.org \
--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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox