linux-ext4 - Re: "tune2fs -I 256" runtime---can it be interrupted safely?

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-id: <20081114082819.GM16005@webber.adilger.int>
Date:	Fri, 14 Nov 2008 01:30:00 -0700
From:	Andreas Dilger <adilger@....com>
To:	"Michael B. Trausch" <mike@...usch.us>
Cc:	linux-ext4@...r.kernel.org
Subject: Re: "tune2fs -I 256" runtime---can it be interrupted safely?

On Nov 11, 2008  11:09 -0500, Michael B. Trausch wrote:
> I do have an ltrace up and online at:
> 
>   http://fd0man.theunixplace.com/tune2fs.ltrace.bz2
> 
> The bzip2 file is 44 MB, and the uncompressed text is _huge_ (about 3
> GB).  This is a two hour run of "tune2fs -I 256 /dev/sdb" on an
> external 250 GB (235 GiB) drive formatted with the ext3 filesystem.
> I aborted the tune2fs run after 2 hours, though I am sure it would have
> run for far longer.  Hopefully, this will give some insight as to what
> the problem might be.

It looks like the code is just spending hours looking at the bitmaps,
but I have no idea why:

ext2fs_test_generic_bitmap(0x1f3d020, 0xfa31e1, 1, 0, 0x1f418b0) = 0
ext2fs_test_generic_bitmap(0x1f3d020, 0xfa31e2, 2, 1, 0x1f418b0) = 0
ext2fs_test_generic_bitmap(0x1f3d020, 0xfa31e3, 4, 2, 0x1f418b0) = 0
ext2fs_test_generic_bitmap(0x1f3d020, 0xfa31e4, 8, 3, 0x1f418b0) = 0
ext2fs_test_generic_bitmap(0x1f3d020, 0xfa31e5, 16, 4, 0x1f418b0) = 0
ext2fs_test_generic_bitmap(0x1f3d020, 0xfa31e6, 32, 5, 0x1f418b0) = 0
ext2fs_test_generic_bitmap(0x1f3d020, 0xfa31e7, 64, 6, 0x1f418b0) = 0
ext2fs_test_generic_bitmap(0x1f3d020, 0xfa31e8, 128, 7, 0x1f418b0) = 0

Possibly the code is very inefficient and is scanning every bit in
the bitmap individually to see if an inode is in use, instead of doing
something fast like just checking an entire bitmap word == 0 and
skipping it.

Unfortunately, ltrace doesn't show what is calling this code, so
a stack trace is needed from GDB.

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@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html