linux-ext4 - Re: [PATCH] e2fsprogs: remove fragment support (V3)

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]

Date:	Mon, 13 Aug 2007 05:59:39 +0800
From:	Coly Li <coyli@...e.de>
To:	Theodore Tso <tytso@....edu>
CC:	linux-ext4@...r.kernel.org, Andreas Dilger <adilger@...sterfs.com>
Subject: Re: [PATCH] e2fsprogs: remove fragment support (V3)

Theodore Tso wrote:
> On Mon, Aug 13, 2007 at 12:33:26AM +0800, Coly Li wrote:
>> Theodore,
>>
>> Another question of mine is for the .po files: Once I modify a string from _(""), should I mofify
>> all the corresponded strings from all .po files ?
> 
> No, don't bother.  Periodically I run "make update-pot" in the po
> directory, which updates the e2fsprogs.pot file, followed by a "make
> update-po" which remove the strings from the .po files.
> 
>> Oops, I thought that issues from my email client. OK, next time when
>> I send patch I will use MIME-PGP.
> 
> Yeah, it's protecting the "-----BEGIN PGP SIGNED MESSAGE-----" header
> line, so it has to add "- " to all lines that begin with "-".  Using
> MIME-PGP means that we use the MIME message boundaries instead.
> 
>>> In addition, for now, I'd like to keep the code which checks the
>>> remaining fragment fields in the inode and superblock since they act
>>> as a safety check to make sure the filesystem is sane and we don't
>>> have garbage there.  The checksum fields will obviate this need, but
>>> keeping the checks there for ext2/ext3 filesystems seem like a good
>>> idea.
>> My idea is:
>> 1) Modify names of related fields of superblock and inode. to avoid others using these field in future.
>> 2) Keep checking code for the modified fields. to make source code robust.
> 
> Comments in the header file is probably enough, I think.  Changing the
> names fields just causes a lot of code churn....
> 
>>> Dropping the union is probably fine, since at this point it looks
>>> pretty clear that both the Hurd and Masix is dead.  But let's do the
>>> cleanups a little at a time, and I'd probably start with just removing
>>> the cruft from the mke2fs options and man pages.
>> Sure, I agree with you :-)
> 
> I did some checking, and it looks like Hurd is still using basic ext2.
> Some grad student did a ext3 driver for Hurd, but it's unstable and
> was apparently last touched in 2005 (the Ph.D. thesis was done in 2003
> iirc).  Sigh...
> 
> Unfortunately, I did some checking and it looks like there are some
> crazy people that are still actively working on a Debian GNU/Hurd
> project.  So let's leave the Hurd stuff in for now.  There may be some
> issues where they will need to drop using ext2 for licensing reasons,
> but let's save this cleanup for later.  It doesn't cost us much to
> leave it in, after all.
> 
Sure, I will only post a patch for removing masix. For hurd, let's wait and see :-)

> 					- Ted

Thanks for your comment :-)
Best regards.

-- 
Coly Li
SuSE PRC Labs


Download attachment "signature.asc" of type "application/pgp-signature" (250 bytes)