linux-kernel - 2.6.18-rc4-mm3 crypto issues with encrypted disks

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

Message-Id: <200609041602.k84G2SYc005390@turing-police.cc.vt.edu>
Date:	Mon, 04 Sep 2006 12:02:28 -0400
From:	Valdis.Kletnieks@...edu
To:	Andrew Morton <akpm@...l.org>,
	Herbert Xu <herbert@...dor.apana.org.au>
Cc:	linux-kernel@...r.kernel.org
Subject: 2.6.18-rc4-mm3 crypto issues with encrypted disks

Sorry for not catching this one earlier..  Sometime between 2.6.18-rc4-mm2
and -mm3, something crept into the git-cryptodev.patch that breaks mounting
encrypted disks.  What I have in /etc/fstab:

/dev/rootvg/crypto1     /crypto/mount_dir    ext3    nodev,nosuid,noexec,noauto,noatime,nodiratime,user,loop,encryption=aes 1 0

This worked in -mm2, and a bisect of -mm patches points git-cryptodev as
the problem, so it's one of the commits in that patch between -mm2 and -mm3.

/sbin/mount is able to set up the loopback, but then gets an error that
no valid superblock was found - which says to me that it's not treating the
passphrase or something the same, and decrypting to something other than what
the -mm2 kernel decrypted the superblock to.

My personal guess as "most likely suspect" (only obvious hit on cryptoloop):

commit d1bc13c88efaa1c1f4f78ac5510297f3187c7f63
Author: Herbert Xu <herbert@...dor.apana.org.au>
Date:   Sun Aug 13 08:45:10 2006 +1000

    [BLOCK] cryptoloop: Use block ciphers where applicable

    This patch converts cryptoloop to use the new block cipher type where
    applicable.  As a result the ECB-specific and CBC-specific transfer
    functions have been merged.

    Signed-off-by: Herbert Xu <herbert@...dor.apana.org.au>

but I have *not* extracted that one GIT commit to test it yet.

Content of type "application/pgp-signature" skipped