From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Date: Thu, 30 Mar 2017 23:36:01 -0700 From: Eric Biggers Subject: Re: [PATCH] fscrypt: Add support for AES-128-CBC Message-ID: <20170331063601.GB32409@zzz> References: <20170330173840.72909-1-david@sigma-star.at> <20170331062149.GA32409@zzz> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20170331062149.GA32409@zzz> To: David Gstir Cc: tytso@mit.edu, jaegeuk@kernel.org, dwalter@sigma-star.at, richard@sigma-star.at, herbert@gondor.apana.org.au, linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org, linux-fscrypt@vger.kernel.org List-ID: On Thu, Mar 30, 2017 at 11:21:49PM -0700, Eric Biggers wrote: > > Something else to consider (probably for the future; this doesn't necessarily > have to be done yet) is that you really only need one essiv_tfm per *key*, not > one per inode. To deduplicate them you'd need a hash table or LRU queue or > something to keep track of the keys in use. > Sorry, I screwed this up. This wouldn't work because the ESSIV key is being derived from the per-file key, not the master key. - Eric