linux-kernel - Re: [PATCH] eCryptfs: infinite loop bug

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: <20120118210400.GB20576@boyd>
Date:	Wed, 18 Jan 2012 15:04:01 -0600
From:	Tyler Hicks <tyhicks@...onical.com>
To:	Linus Torvalds <torvalds@...ux-foundation.org>
Cc:	dragonylffly <dragonylffly@....com>, john.johansen@...onical.com,
	dustin.kirkland@...zang.com, ecryptfs@...r.kernel.org,
	linux-fsdevel@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] eCryptfs: infinite loop bug

On 2012-01-18 12:49:17, Linus Torvalds wrote:
> Hmm.
> 
> There are *two* cases where we do that "total_remaining_bytes"
> calculation. The same bug seems to exist both in ecryptfs_read() and
> ecryptfs_write().

ecryptfs_read() is ifdef'ed out, and has been for years, so I'll just go
ahead and kill that function for good.

> 
> Possibly only the ecryptfs_write() one leads to an endless loop, but
> the read one looks suspicious too.
> 
> Also, what protects things against this just being one nasty DoS
> attack - even if the code is fixed to not be an endless loop, it looks
> like a trivial "truncate()" can be used to generate a *practically*
> infinite write stream. At the very least, this should be KILLABLE. Or
> did I mis-read the code?

I think you're right. I'll start off with making it killable and then
see if there is anything else we can do.

Tyler

> 
> Tyler, Dustin, others - comments? This looks nasty.
> 
>                      Linus.
> 
> 2012/1/17 dragonylffly <dragonylffly@....com>:
> > Hi,
> >   There is an infinite loop bug in eCryptfs, to make it present,
> > just truncate to generate a huge file (>= 4G) on a 32-bit machine under
> > the plain text foleder mounted with eCryptfs, a simple command
> > 'truncate -s 4G dummy' is enough. Note: 4GB is smaller than 4G,
> > therefore the following command 'truncate -s 4GB dummy' will not
> > trigger this bug. The bug comes from a data overflow, the patch below fixes
> > it.
> >
> > Cheers,
> > Li Wang
> >
> > ---
> >
> > signed-off-by: Li Wang <liwang@...t.edu.cn>
> >                Yunchuan Wen (wenyunchuan@...inos.com.cn)
> >
> > --- read_write.c.orig 2012-01-18 10:39:26.000000000 +0800
> > +++ read_write.c 2012-01-18 19:48:41.484196221 +0800
> > @@ -130,7 +130,7 @@
> >    pgoff_t ecryptfs_page_idx = (pos >> PAGE_CACHE_SHIFT);
> >    size_t start_offset_in_page = (pos & ~PAGE_CACHE_MASK);
> >    size_t num_bytes = (PAGE_CACHE_SIZE - start_offset_in_page);
> > -  size_t total_remaining_bytes = ((offset + size) - pos);
> > +  loff_t total_remaining_bytes = ((offset + size) - pos);
> >
> >    if (num_bytes > total_remaining_bytes)
> >     num_bytes = total_remaining_bytes;
> >
> >
> >
> --
> To unsubscribe from this list: send the line "unsubscribe ecryptfs" in
> the body of a message to majordomo@...r.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

Download attachment "signature.asc" of type "application/pgp-signature" (837 bytes)