Markus Schuster wrote: > 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? On a core2duo machine when I use these parameters | -c aes-cbc-essiv:sha256 -h sha256 -s 128 -y | -c aes-lrw-benbi -h sha256 -s 256 -y and encode around 5GB of /dev/zero, both cores are complety used. But lrw is around 8% faster than cbc-essiv (ca. 75 MB/s with lrw and ca. 69 MB/s with cbc-essiv). Regards, Till --------------------------------------------------------------------- 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
