LWN: Comments on "Next-interrupt prediction"

Next-interrupt prediction

physkets — Fri, 20 Oct 2017 11:31:57 +0000

Such machines would be Turing-overcomplete

Next-interrupt prediction

nix — Fri, 08 Sep 2017 20:28:06 +0000

If you know of a processor equipped with a wormhole peripheral, you can do much more useful stuff with it -- creating Turing oracles, for instance.

Next-interrupt prediction

mips — Thu, 04 Feb 2016 17:50:05 +0000

I also wonder what the power cost of maintaining all this accounting is, relative to doing no prediction (or very naive prediction, like the next-timer approach mentioned).

Next-interrupt prediction

clicea — Thu, 04 Feb 2016 02:13:14 +0000

I do wonder, though, how realiable is this information and how resistant to outliers. See Anscombe's quartet for examples on how can you be mislead using only means and standard deviations.

Next-interrupt prediction

david.a.wheeler — Sun, 31 Jan 2016 20:05:51 +0000

> > Unfortunately, wormhole peripherals that can read information from the future are expensive, so the vast majority of processors are not equipped with them.

> Quote of the week :)

Indeed. Or is that quote of next week? :-).

Next-interrupt prediction

tdalman — Sun, 31 Jan 2016 10:32:36 +0000

> Unfortunately, wormhole peripherals that can read information from the future are expensive, so the vast majority of processors are not equipped with them.

Quote of the week :)