[Cryptography] Speculation considered harmful?

Henry Baker hbaker1 at pipeline.com
Tue Jan 9 00:00:17 EST 2018


At 03:35 AM 1/8/2018, John Levine wrote:
>In article <7f4174cb-b842-1314-587a-dd32711a81bf at symas.com> you write:
>>> One of them is VLIW, or "Very Long Instruction Word," which exploits
>>> deliberately explicit instruction level parallelism rather than implicit
>>> (speculative) instruction parallelism.
>>
>>Intel EPIC -> Itanium -> nobody liked that path.
>
>There is room for debate about what was wrong with Itanium.
>
>I know the guys who invented VLIW.
>
>It speculates all over the place, with part of the idea being that some of the work the long instructions do is thrown away if it turns out that it's on a path that turns out not to be taken.
>
>I have no idea whether its flavor of speculation can be used for the same kind of attacks.

You knew the guys who invented music boxes ?

You knew the MIT Whirlwind guys and Maurice Wilkes ?  ;-)

https://en.wikipedia.org/wiki/Microcode#History

They invented the first VLIW architectures, although their Very Long (microcode) Instruction Words couldn't be dynamically written.

I knew the Multflow VLIW company, but I would say that there were a lot of sophisticated things going on that went far beyond the simple idea of "very long word instructions".

The concept of "speculation" only becomes important when you have such an excess of hardware resources, that you bend over backwards trying to give them something to do -- the computer science version of Parkinson's Law.

https://en.wikipedia.org/wiki/Parkinson%27s_law

The early time-sharing systems research concluded that there is a hard limit to how much "efficiency" you can get out of shared resources, after which the competing processes damage each other more than they gain.

Of course, computer scientists could simply have asked any economist (or even a mathematician or an army general): they have been aware for hundreds (thousands) of years that *slack resources* ("reserves" in army lingo) have significant value even when not currently being "used"/"consumed".

Of course, army reserves are constantly training, which might be the equivalent of computer hardware constantly running hardware diagnostic programs to make sure that nothing in the hardware is failing.



More information about the cryptography mailing list