On Tue, Sep 18, 2007 at 01:00:43AM +0200, Markus Schuster wrote: > Arno Wagner wrote: > > > For many practical applications, ESSIV and LRW should be equally > > secure. Note that with both you have absolutely no leaking of the > > contens of the plain data, just some leakage about changes. > > I haven't really noticed LRW has made it's way in the vanilla kernel? I > remember LRW from Clemen's tries to push it in the mainline linux kernel > and the problems he had. > Now that LRW is in the kernel, I have a question regarding it's > implementation: As far as I know, LRW can be parallelized and so make use > of more CPU cores, in contrast with CBC, which can only use one CPU. Is > this correct for the current implementation? I have no idea whether it is done. But I would suspect if there is parallelism it is on sector-level, since that would be far easier to do and in most cases faster. Arno -- Arno Wagner, Dipl. Inform., CISSP --- CSG, ETH Zurich, wagner@xxxxxxxxxxxxxx GnuPG: ID: 1E25338F FP: 0C30 5782 9D93 F785 E79C 0296 797F 6B50 1E25 338F ---- Cuddly UI's are the manifestation of wishful thinking. -- Dylan Evans If it's in the news, don't worry about it. The very definition of "news" is "something that hardly ever happens." -- Bruce Schneier --------------------------------------------------------------------- dm-crypt mailing list - http://www.saout.de/misc/dm-crypt/ To unsubscribe, e-mail: dm-crypt-unsubscribe@xxxxxxxx For additional commands, e-mail: dm-crypt-help@xxxxxxxx
