On 04/11/2017 10:18 AM, Max Reitz wrote: > Hm, yeah, although you have to keep in mind that the padding is almost > pretty much the same as the the data bits we need, effectively doubling > the size of the L2 tables: > > padding = 2^{n+2} - 2^{n+1} - 64 (=2^6) > = 2^{n+1} - 64 > > So that's not so nice, but if it's the only thing we can do...
Or we mix-and-match your ideas: since our subclusters are ternary encoding, if you want 128 subclusters, instead of asking for 256 bits, you ask for ld(3^128) = 203 bits, then use 11 padding bits of your original 64, and you have something that still fits in 256 bits instead of 512... Okay, just kidding. Anything larger than 32 subclusters does get awkward fast. -- Eric Blake, Principal Software Engineer Red Hat, Inc. +1-919-301-3266 Virtualization: qemu.org | libvirt.org
signature.asc
Description: OpenPGP digital signature