linux-kernel - Re: BUG: Failure to send REQ_FLUSH on unmount on ext3, ext4, and FS in general

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

Message-ID: <20110523173350.GA10554@infradead.org>
Date:	Mon, 23 May 2011 13:33:50 -0400
From:	Christoph Hellwig <hch@...radead.org>
To:	Jan Kara <jack@...e.cz>
Cc:	Alex Bligh <alex@...x.org.uk>, linux-kernel@...r.kernel.org,
	Christoph Hellwig <hch@...radead.org>,
	Andrew Morton <akpm@...ux-foundation.org>,
	Andreas Dilger <adilger.kernel@...ger.ca>,
	Theodore Ts'o <tytso@....edu>
Subject: Re: BUG: Failure to send REQ_FLUSH on unmount on ext3, ext4, and FS
 in general

On Mon, May 23, 2011 at 07:29:06PM +0200, Jan Kara wrote:
>   Yes, but that's rather a deficiency in default mount options of ext3
> which is kept for backward buggy-for-performance compatibility. Anywone who
> seriously cares about the data should use barrier=1 and BTW SUSE or RH
> distros change the default to be barrier=1. Anyway, this is a seperate
> issue.

It really should be changed.  The previous (bad) excuse was that the
ordering barrier code was too much overhead.  Making a filesystem
non-safe by default is already a bad sin, but having the code to make
it safe around and not enabling it is plain criminal.

> > Would there be any interested in these patches if I cooked them up,
> > or did they die because of opposition before rather than apathy?
>   I guess you might come with some proposal and post it to linux-fsdevel
> (include Al Viro and Christoph Hellwig in CC) and see what happens...

There's no way to make it generic.  Fortunately adding support to a
filesystem is generally trivial, take a look at my recently added
cache flushing support for hfsplus for example.

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