From: Richard Weinberger Subject: Re: [PATCH] fscrypto: make XTS tweak initialization endian-independent Date: Tue, 4 Oct 2016 10:46:54 +0200 Message-ID: <6c9b32ef-1e63-f721-1d7f-b0f1e0f2d1ca@nod.at> References: <1475258329-146528-1-git-send-email-ebiggers@google.com> <20161003180340.GA54410@google.com> Mime-Version: 1.0 Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: 7bit Cc: linux-fsdevel , linux-ext4@vger.kernel.org, linux-f2fs-devel@lists.sourceforge.net, Theodore Ts'o , jaegeuk@kernel.org, David Gstir To: Eric Biggers Return-path: In-Reply-To: <20161003180340.GA54410@google.com> Sender: linux-fsdevel-owner@vger.kernel.org List-Id: linux-ext4.vger.kernel.org Eric, On 03.10.2016 20:03, Eric Biggers wrote: >>> { >>> - u8 xts_tweak[FS_XTS_TWEAK_SIZE]; >>> + struct { >>> + __le64 index; >>> + u8 padding[FS_XTS_TWEAK_SIZE - sizeof(__le64)]; >>> + } xts_tweak; >> >> While we are here, wouldn't it make sense to rename the variable to "iv"? >> In aes-xts mode the IV is used as tweak. But it is still an IV and passed >> as IV parameter to the crypto API. >> >> Especially when other cipher modes are used this is confusing. >> > > Good idea --- I agree that "iv" is a better name, so as to not tie the code to > XTS specifically. But I think the renaming should be a separate patch. Sure. I can do that. > Also, currently this code *is* only supposed to be used for XTS. There's a bug > where a specially crafted filesystem can cause this code path to be entered with > CTS, but I have a patch pending in the ext4 tree to fix that. David and I are currently working on UBIFS encryption and we have to support other cipher modes than XTS. So, keeping fscrypto as generic as possible would be nice. :-) Thanks, //richard