Re: [LTP] [PATCH] Add new testcases for ext4 new features - V3

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On 2009-10-26, at 07:26, Subrata Modak wrote:
We had recently added EXT4 testcases to LTP. As the author of these
tests mentioned, we saw 2 such failures in subdirectory limit test:

Ext4 subdir limit test
ext4-subdir-limit 0 TINFO : Num of dirs to create: 65537, Dir name
len: short name                        Parent dir: mnt_point, Block
size: 1024
mke2fs 1.41.9 (22-Aug-2009)
tune2fs 1.41.9 (22-Aug-2009)
Setting test filesystem flag
/dev/VG1_EXT4/LV1_EXT4: The test_fs flag is set (and ext4 is available).
CLEARED.

You don't need to set the test_fs flag for modern ext4.

/dev/VG1_EXT4/LV1_EXT4: clean, 10/8847360 files, 2297019/141557760
blocks
dumpe2fs 1.41.9 (22-Aug-2009)
ext4-subdir-limit    1  TPASS  :  ext4 subdir limit test pass
ext4-subdir-limit 0 TINFO : Num of dirs to create: 65537, Dir name
len: long name                         Parent dir: mnt_point, Block
size: 1024
mkdir: No space left on device
ext4-subdir-limit    2  TFAIL  :  failed to create directories - 19524
ext4-subdir-limit 0 TINFO : Num of dirs to create: 65537, Dir name len: short name Parent dir: mnt_point/sub, Block
size: 1024
mke2fs 1.41.9 (22-Aug-2009)
tune2fs 1.41.9 (22-Aug-2009)
Setting test filesystem flag
/dev/VG1_EXT4/LV1_EXT4: The test_fs flag is set (and ext4 is available).
CLEARED.
/dev/VG1_EXT4/LV1_EXT4: clean, 12/8847360 files, 2297037/141557760
blocks
dumpe2fs 1.41.9 (22-Aug-2009)
ext4-subdir-limit    3  TPASS  :  ext4 subdir limit test pass
ext4-subdir-limit 0 TINFO : Num of dirs to create: 65537, Dir name len: long name Parent dir: mnt_point/sub, Block
size: 1024
mkdir: No space left on device
ext4-subdir-limit    4  TFAIL  :  failed to create directories - 19483
ext4-subdir-limit 0 TINFO : Num of dirs to create: 65537, Dir name
len: short name                        Parent dir: mnt_point, Block
size: 2048

Is this a known issue for EXT4 ?

Creating many long filenames (255 bytes) in a 1kB block filesystem is a
known issue that is not expected to be fixed (at least I'm not aware of
any work being done in this area).  A vast majority of filesystems
today are formatted with 4kB blocks (due to disk size) and the overhead of
managing 1kB block filesystems makes these only marginally used, and it
is also very rare to have applications with such large filenames.

We also uncovered that the tests gives a decent 50% lines and over 61%
function coverage for EXT4 code. Please find the individual coverage
under EXT4 attached.

Interesting to see this.

Following the various system information on which we ran the test:
1) uname -a
Linux 2.6.31 #1 SMP Tue Oct 20 22:57:16 IST 2009 i686 i686 i386
GNU/Linux

2) grep -i ext4 .config
CONFIG_EXT4_FS=m
CONFIG_EXT4DEV_COMPAT=y
CONFIG_EXT4_FS_XATTR=y
CONFIG_EXT4_FS_POSIX_ACL=y
CONFIG_EXT4_FS_SECURITY=y

3) mkfs.ext4 -V
mke2fs 1.41.9 (22-Aug-2009)
       Using EXT2FS Library version 1.41.9
e2fsprogs-1.41.9
e2fsprogs-libs-1.41.9

4) LTP used:
Unreleased CVS repository. See the URL below to download:
http://ltp.sourceforge.net/documentation/how-to/ltp.php#_3.1,

You can also also directly review the attached patch(which added this
test cases):
26_10_2009-(Miao-Xie<miaox@xxxxxxxxxxxxxx>)- Add_new_testcases_for_ext4_new_features_V3.patch.regenerated,

5) I used a single 680GB LVM partition to test this.

Kindly let us know your thoughts on this failures.

Cheers, Andreas
--
Andreas Dilger
Sr. Staff Engineer, Lustre Group
Sun Microsystems of Canada, Inc.

--
To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[Index of Archives]     [Reiser Filesystem Development]     [Ceph FS]     [Kernel Newbies]     [Security]     [Netfilter]     [Bugtraq]     [Linux FS]     [Yosemite National Park]     [MIPS Linux]     [ARM Linux]     [Linux Security]     [Linux RAID]     [Samba]     [Device Mapper]     [Linux Media]

  Powered by Linux