Date: Mon, 8 Sep 2014 04:52:21 GMT From: John-Mark Gurney <jmg@FreeBSD.org> To: Perforce Change Reviews <perforce@FreeBSD.org> Subject: PERFORCE change 1199754 for review Message-ID: <201409080452.s884qLgG029413@skunkworks.freebsd.org>
index | next in thread | raw e-mail
http://p4web.freebsd.org/@@1199754?ac=10 Change 1199754 by jmg@jmg_carbon2 on 2014/09/03 17:48:24 after testing, I have confirmed that _GCM does not suffer the same problem that _ICM does, and it can be used on arbitrary byte boundaries... other minor formating issues wrt bits... Sponsored by: FreeBSD Foundation Affected files ... .. //depot/projects/opencrypto/share/man/man7/crypto.7#2 edit Differences ... ==== //depot/projects/opencrypto/share/man/man7/crypto.7#2 (text+ko) ==== @@ -59,7 +59,7 @@ .It IV size : 12 .It Block size : -1 (aesni), 16 (software) +1 .It Key size : 16, 24 or 32 .El @@ -91,9 +91,9 @@ This is similar to what most people call counter mode, but instead of the counter being split into a nonce and a counter part, then entire nonce is used as the initial counter. -This does mean that if a counter is required that rolls over at 32bits, +This does mean that if a counter is required that rolls over at 32 bits, the transaction need to be split into two parts where the counter rolls over. -The counter incremented as a 128bit big endian number. +The counter incremented as a 128-bit big endian number. .It Dv CRYPTO_AES_XTS .Bl -tag -width "Block size :" -compact -offset indent .It IV size :help
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201409080452.s884qLgG029413>