[Cryptography] Crypto for optimistic transactions ?

[Jeremy Stanley]

On 2018-01-06 06:55:25 +0000 (+0000), Howard Chu wrote:
> Tom Mitchell wrote:
> > A third landmark goal is to design hardware that does not take
> > the 5-35% perf hit.  Whoever solves this first will sell a lot
> > of hardware including motherboard and DRAM makers. But this will
> > put a chill on sales for months.
> 
> That 5-35% number is the effect of the patches to mitigate
> Meltdown, which primarily affects Intel (and also ARM's
> Cortex-A75). AMD clearly demonstrates that there's a right way to
> handle things, since their chips don't have this problem.
[...]

Also amusing (to me at any rate) is that OpenBSD preemptively
implemented basically the same mitigations a decade ago, in the wake
of the 2006 Intel Core 2 errata. This rant from Theo is remarkably
prophetic:

    https://marc.info/?l=openbsd-misc&m=118296441702631&w=2

Now fast-forward to modernity:

    https://bsd.network/@phessler/99291274119329309

-- 
Jeremy Stanley
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 963 bytes
Desc: not available
URL: <http://www.metzdowd.com/pipermail/cryptography/attachments/20180106/219a6e78/attachment.sig>