Re: Don't use S-boxes!

From: Stefan Tillich (stefanti_at_gmx.at)
Date: 11/24/04 

Date: Wed, 24 Nov 2004 08:30:15 +0100

karl_m@acm.org schrieb:
> D. J. Bernstein wrote:
>
>>karl malbrain wrote:
>>
>>>I don't think that pre-loading the S-BOX into LEVEL-ONE CACHE will
>>>help in the attacks you demonstrate BECAUSE THEY DON'T FIT IN THE
>>>FIRST PLACE.
>>
>>False. For example, the Athlon's level-1 data cache has 65536 bytes,
>
> as
>
>>stated in the sixth paragraph of the ``Load-timing variability''
>
> section
>
>>of my paper. The most extreme AES encryption tables I've seen are
>
> 8192
>
>>bytes.
>
>
> Where I come from this response would be considered VULGAR.
>
>
>>There are several reasons that preloading the tables doesn't stop
>
> timing
>
>>attacks---as the same section points out in considerable detail---but
>>the size of the L1 cache is not one of those reasons.
>
>
> But SECTION FOUR clearly illustrates the MULTIPLEX of both the
> ASSOCIATIVITY BITS and the BANK SELECT BITS with the LINE SELECT BITS
> in the CACHE ADDRESS SPACE. What gives?

OT: All uppercase words are considered as shouted, which does not
conform to the Netiquette guidelines (RFC 1855). I think most of the
readers of this forum will be able to grasp the meaning of your
statements without you shouting half of your postings.

No offence.

Best regards
Stefan Tillich

Relevant Pages

  • Re: [RFC] Limit the size of the pagecache
    ... I have gotten repeatedly requests to be able to limit the pagecache. ... It may be useful to limit the size of the page cache for various reasons ... Insure rapid turnaround of pages in the cache. ... Send instant messages to your online friends http://au.messenger.yahoo.com - ...
    (Linux-Kernel)
  • Re: which OutputStreams are buffered?
    ... I guess one of the reasons is that it can be very difficult to ... implement an API that make it 100% sure the data is at location that ... Cache in RAID controllers, cache in disk drives, ... in the war against terrorism and a new information agency would help ...
    (comp.lang.java.programmer)
  • Re: SP Execution Very Slow
    ... --dbcc freeproccache -- this one drops proc cache in all db. ... DECLARE @intDBID INTEGER ... DBCC FLUSHPROCINDB ... >> There can be other reasons. ...
    (microsoft.public.sqlserver.programming)
  • Re: Migration away from mvbase?
    ... and Cache, and I'm heavily bent toward Cache in this case. ... well as the advantages / disadvantages you found with migrating. ... Main reasons for looking at migrating: ...
    (comp.databases.pick)