More importantly, the exec says that Intel has finally identified the root cause of reboot issues on Broadwell and Haswell processors. It also advised customers of certain systems (a full list is here) to stop deploying the patches until the updated patches are released. In fact, an early version of the updated solution has begun rolling out to industry partners initially.
Shenoy then confirmed that the computer reboots were experienced on systems powered by Ivy Bridge-, Sandy Bridge-, Skylake-, and Kaby Lake-based platforms, among others.
Each time news trickles out about just how insane this unprecedented industry event really is, cloud vendors and data center managers consider buying servers from AMD or ARM vendors the next time their systems need to be replaced.
Shenoy also posted updates on the performance tests they were conducting in data centers. Specifically, it is causing unexpected reboot issues. Intel is, however, working on a new firmware update to address the random reboot issues. But the impact to older CPUs is known to be harder than on newer chips, and we're still waiting for patches to roll out for the major variants (there are three subtypes of Meltdown and Spectre attacks as of this writing) and to see how users are going to be impacted in common workloads.
Torvalds flagged the possibility of an extra release candidate last week, with the caveat that "it obviously requires this upcoming week to not come with any huge surprises" after "all the Meltdown and Spectre hoopla" made his job rather more complicated in recent weeks.
Intel first acknowledged the problem more than a week ago, saying chips in the company's lines called Broadwell and Haswell were causing problems after receiving updates.
"We continue to urge all customers to vigilantly maintain security best practice and for consumers to keep systems up-to-date", Shenoy continued.