On Wed, 5 Sep 2018, Andi Kleen wrote: > Please if you repost include plenty of performance numbers for multi > threaded workloads. It's ridiculous to even discuss this without them.
Talking about ridiculous ... I find it a bit sad that Intel has let this be unfixed for 3/4 years in linux; that doesn't really signal deep dedication to customer safety. Have any STIBP patches been even submitted? This is not the same situation as IBRS which was mostly ignored -- there we have retpolines to protect the kernel, and it's debatable whether it's exploitable on SKL at all. Ignoring IBPB and STIBP is keeping the system plain vulnerable to user-user attacks, and us not providing users with possibiliy to easily mitigate, is a bit embarassing in my eyes. Thanks, -- Jiri Kosina SUSE Labs