From mboxrd@z Thu Jan 1 00:00:00 1970 From: Dave Hansen Date: Sat, 08 Dec 2018 01:11:58 +0000 Subject: Re: [RFC v2 00/13] Multi-Key Total Memory Encryption API (MKTME) Message-Id: <5862ff39-e4ab-2a04-95be-84d2e8b67120@intel.com> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit List-Id: References: <1544147742.28511.18.camel@intel.com> In-Reply-To: To: Andy Lutomirski , kai.huang@intel.com Cc: "Kirill A. Shutemov" , James Morris , Peter Zijlstra , keyrings@vger.kernel.org, Matthew Wilcox , Thomas Gleixner , Linux-MM , David Howells , LSM List , Dan Williams , X86 ML , "H. Peter Anvin" , Ingo Molnar , "Sakkinen, Jarkko" , Borislav Petkov , Alison Schofield , Jun Nakajima On 12/7/18 3:53 PM, Andy Lutomirski wrote: > The third problem is the real show-stopper, though: this scheme > requires that the ciphertext go into predetermined physical > addresses, which would be a giant mess. There's a more fundamental problem than that. The tweak fed into the actual AES-XTS operation is determined by the firmware, programmed into the memory controller, and is not visible to software. So, not only would you need to put stuff at a fixed physical address, the tweaks can change from boot-to-boot, so whatever you did would only be good for one boot.