More details about mitigations for the CPU Speculative Execution issue

Re:More details about mitigations for the CPU Speculative Ex... by leonardinius on Thursday January 04, 2018 @04:45PM (#16074352)

And now I have to wonder about negligible performance implications.

Does it mean Google applied point-and-shoot type of fixes in several areas, e.g. OS kernel and hypervisor? Most likely.

Or. The speculative execution did not provide meaningful performance benefits in the first place? At least for Google workloads? If so - why all this extra complexity?

[ Parent | on HN ]

Re:More details about mitigations for the CPU Speculative Ex... by jpatokal on Thursday January 04, 2018 @05:14PM (#16074585)

The mitigations don't involve disabling speculative execution entirely, only adding safeguards to how it's done.

[ Parent | on HN ]

Re:More details about mitigations for the CPU Speculative Ex... by Sir_Cmpwn on Thursday January 04, 2018 @05:30PM (#16074714)

Speculative execution definitely does provide meaningful performance benefits, but like many things it only really shows itself clearly in a few algorithms. Branch prediction failure is a fairly well known source of performance issues.

[ Parent | on HN ]