Hacker News Reader

cws125on Intel Issues Updates to Protect Systems from Security Exploits

It appears the retpoline fixes don't work in Skylake or later (it's smart enough to speculate out of it?) and will require new support for IBRS/IBPB in the microcode to mitigate.

From: https://lkml.org/lkml/2018/1/4/615

cws125on LLVM patch to fix half of Spectre attack

Just as a FYI, according to:

* https://lkml.org/lkml/2018/1/4/432 * http://xenbits.xen.org/gitweb/?p=people/andrewcoop/xen.git;a...

It appears that Skylake and later can actually predict retpolines? Some hardware features called IBRS, IBPB, STIBP (not a lot of details on this are out there) are supposedly coming in a microcode update.

cws125on A Most Peculiar Test Drive

Both Broder and Musk agree that the car showed 32 miles of range, and that the car had made it 51 miles. But he would have known, before he left Norwich, that the distance to Milford was 61 miles. How Broder reasonably expect he would get there?

Broder claims: "Tesla’s experts said that pumping in a little energy would help restore the power lost overnight as a result of the cold weather, and after an hour they cleared me to resume the trip to Milford."

In other words, it sounds like he thought that the 65 miles of range he lost overnight would be restored by charging it a little bit more at Norwich. He says he just couldn't make it to the next charging station by the time he realized this clearly wasn't going to work.

cws125on A Most Peculiar Test Drive

No, I think overnight was at 400 miles. The Milford Supercharge was at roughly 322 miles according to the graph. Broder says he traveled 79 miles before spending the night at Groton, so that's at roughly 400 miles.