-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Fri, Jan 05, 2018 at 10:33:45AM -0600, Nicholas Geovanis wrote: > On Fri, Jan 5, 2018 at 6:55 AM, <to...@tuxteam.de> wrote: > > (mainframes of that time had at least VM, possibly > > speculative prefetch). > > Is it correct to call branch prediction the same as speculative execution?
I'm far from a processor expert myself but yes, that sounds plausible: the memory unit is prefeching stuff on the speculation that the branch will go one way, to keep the pipeline full. When things go the other way... > If so, then "yes" they had it, but I don't honestly know if that's correct. > Pipeline rewinding was necessary on prediction failure, etc, similarly. ... this happens. I guess that this, in combination with a cache, could be enough for the kind of timing attacks demonstrated by Spectre. But I'm speculating myself here :) Cheers - -- tomás -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.12 (GNU/Linux) iEYEARECAAYFAlpPq9QACgkQBcgs9XrR2kZPhQCeNpZYsjInyrAh5HNPqTkJ6w33 s9gAn1th2QyrDsjqu/dOlZcOjRFTC3Mo =dV+r -----END PGP SIGNATURE-----
