Hi! I am new here, I just subscribed. Greetings! My name is Henrik Theiling and I am a programmer in a small and cosy company in Saarbrücken in south-west Germany. Our interest is chiefly in timing analysis and optimisation of real-time embedded systems. My personal interest in crypto stuff is mainly as a user to protect my data, although I do like to think about security and crypto concepts and implementations. The current story everyone is talking about (even here, it seems!), that RAM remembers the bits after the power is off (which is old knowledge, but the video impressively showed how bad physical access to a secured machine really is) brought back my programmer's interest, so I downloaded cryptsetup-1.0.5 source code yesterday in order to read in it a bit. Just curious how it all fits together internally that I use regularly. But the tarball is probably outdated, right -- is there a CVS or SVN version somewhere? I found a potential denial of service (SIGSEGV), but I'd like to confirm with the current code. Arno Wagner wrote: > On Fri, Feb 22, 2008 at 11:45:14AM -0800, Bill Broadley wrote: > > Do today's CPUs allow for pinning a small amount of data in cache? Say 16 > > bytes or whatever is needed for an encryption key? > > No. There is no structure for that. Caches do not work this way. Some embedded CPUs have cache locking or scratchpad RAM, but not x86 and probably other desktop CPUs. OTOH, would anyone want to shut off a portion of the cache for crypto keys? Hmm, maybe I would. :-) **Henrik --------------------------------------------------------------------- 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
