RIP Hyper-Threading? ChromeOS axes key Intel CPU feature over data-leak flaws – Microsoft, Apple suggest snub

Plug pulled on SMT tech as software makers put security ahead of performance

Analysis In conjunction with Intel's coordinated disclosure today about a family of security vulnerabilities discovered in millions of its processors, Google has turned off Hyper-Threading in Chrome OS to fully protect its users.

Meanwhile, Apple, Microsoft, IBM's Red Hat, QubesOS, and Xen advised customers that they may wish to take similar steps.

The family of flaws are dubbed microarchitecture data sampling (MDS), and Chipzilla's official advisory is here, along with the necessary microcode updates to mitigate the data-leaking vulnerabilities and list of affected products. Installing these fixes and disabling Intel's Hyper-Threading feature is a sure fire way to kill off the bugs, though there may be a performance hit as a result.


Hyper-Threading is Intel's implementation of simultaneous multithreading (SMT), a technique for splitting a single physical processor core into two virtual cores known as hardware threads. It's supposed to improve performance by allowing two software threads to run simultaneously through each physical core, sharing available resources on the silicon as needed. This means one physical core can juggle two threads, either in the same application or two separate applications, at the same time, improving throughput. Some workloads benefit from this, some are hindered or see no gain. You mileage may vary.

However, one thing it does bring into the mix is the risk that side-channel surveillance techniques, such as MDS, may be able to break hardware thread isolation, and access sensitive data it shouldn't be able to see. In other words, one thread can snoop on the memory accesses of another thread sharing the same physical CPU core, and lift passwords, keys, and other secrets, potentially.

Really, today's chip flaw disclosures cover a group of design blunders: ZombieLoad (CVE-2018-12130) can be exploited by malware or rogue users on a vulnerable system to potentially steal browser histories, website content, user keys, passwords, and system-level secrets, such as disk encryption keys from other parts of memory. We're told it can work across CPU protection rings and process boundaries, and against cloud and on-premises virtual machines and trusted execution environments. Proof-of-concept exploit code is available to try it out for yourself.

There's also RIDL and Fallout (CVE-2018-12126, CVE-2018-12127, CVE-2019-11091) that can be exploited to steal confidential info from memory.

Mitigating these security oversights in Intel's chips will require microcode updates to be installed, and operating system and hypervisor patches to utilize them, so check your OS vendor, and system manufacturer if needed, for new software and install it as soon as you're able. These fixes may introduce a performance hit depending on what kind of programs you're running.

You can opt to turn off Hyper-Threading to fully neutralize the threat, though you may want to weigh up if it's worth the performance cost by testing your applications with the feature on and off.


Google said it is disabling Hyper-Threading by default in Chrome OS 74, citing security concerns, and noting that Chrome OS 75 will have additional mitigations.

"The decision to disable or enable Hyper-Threading is a security versus performance tradeoff," said the web giant's people in a vulnerability notice. "With Hyper-Threading disabled, Intel CPUs may experience reduced performance, which varies depending on the workload. But, with Hyper-Threading enabled, users could execute code, such as by visiting a website or running an Android app, that exploits MDS to read sensitive memory contents."

Google has further details on how it's handling the bugs, from its client applications to cloud services, right here.

BSD land

The OpenBSD community, for one, came to that conclusion last year when it disabled Hyber-Threading in OpenBSD 6.4. In response to past Intel processor vulnerabilities (TLBleed and L1TF) that showed Hyper-Threading to be a risk, OpenBSD leader Theo de Raadt observed that Hyper-Threading is fundamentally broken because shares resources between two CPU instances without assuring secure isolation.

"DISABLE HYPERTHREADING ON ALL YOUR INTEL MACHINES IN THE BIOS," he said in a mailing list post at the time.


Apple has released macOS Mojave 10.14.5 to address MDS attacks via JavaScript and Safari. But it says a comprehensive fix requires turning off Hyper-Threading, which comes with a potentially substantial performance cost.

"Full mitigation requires using the Terminal app to enable an additional CPU instruction and disable hyper-threading processing technology," Apple warned in its advisory. "This capability is available for macOS Mojave, High Sierra, and Sierra in the latest security updates and may reduce performance by up to 40 per cent, with the most impact on intensive computing tasks that are highly multithreaded."

Unfortunately for Apple customers with older Macs, Intel has not made microcode fixes available for Mac models from 2010 or earlier.


Microsoft in its MDS threat guidance does not take a firm stand but notes, "To be fully protected, customers may also need to disable Hyper-Threading." The Windows giant has released operating system updates to mitigate Intel's design flaw in conjunction with necessary microcode updates – see the aforementioned link.

Red Hat

Red Hat includes a link to disabling Hyper-Threading in its advisory without making a recommendation one way or another. Its Hyper-Threading (SMT) security page notes, "Various microprocessor flaws have been discovered recently. Certain issues require SMT be disabled in order to more fully mitigate the issue."

The enterprise Linux slinger has more technical notes here and here on the cause and effects – or you can check out the vid below. Other Linux distros should be rolling out their fixes, too. Here's the state of play with Ubuntu and Debian, for instance.

Youtube Video

Google Cloud only recommends disabling Hyper-Threading for Compute Engine users "if you are using Container Optimized OS (COS) as your Guest OS and you are running untrusted, multi-tenant workloads in your virtual machine." It makes a similar recommendation for those running untrusted code on multi-tenant services within Kubernetes Engine.

Xen, which makes a hypervisor used by AWS (advisory) and other cloud providers others, has issued an advisory that details the risks of Hyper-Threading while refusing to disable the technology by default because doing so would be too disruptive. Mitigations and fixes are available from the aforementioned link.

"Leakage of data from Xen or other guests can only prevented entirely by disabling hyper-threading (if available and active in the BIOS), and by applying the patches to Xen," its advisory stated.

Qubes, which relies on Xen for virtualization, says much the same.

Intel is fine with its technology, and leaves the decision to disable Hyper-Threading to its industry partners.

"Intel is not recommending disabling HT," a company spokesperson told The Register in an email.

"It’s important to understand that disabling SMT/HT does not alone provide protection against MDS, and doing so may impact workload performance or resource utilization that can vary depending on the workload.

"After systems are updated, there are some cases where additional considerations may apply. Our software partners will provide guidance that can help customers make the right decisions for their systems and the workloads critical to their needs." ®

Other stories you might like

  • A peek into Gigabyte's GPU Arm for AI, HPC shops
    High-performance platform choices are going beyond the ubiquitous x86 standard

    Arm-based servers continue to gain momentum with Gigabyte Technology introducing a system based on Ampere's Altra processors paired with Nvidia A100 GPUs, aimed at demanding workloads such as AI training and high-performance compute (HPC) applications.

    The G492-PD0 runs either an Ampere Altra or Altra Max processor, the latter delivering 128 64-bit cores that are compatible with the Armv8.2 architecture.

    It supports 16 DDR4 DIMM slots, which would be enough space for up to 4TB of memory if all slots were filled with 256GB memory modules. The chassis also has space for no fewer than eight Nvidia A100 GPUs, which would make for a costly but very powerful system for those workloads that benefit from GPU acceleration.

    Continue reading
  • GitLab version 15 goes big on visibility and observability
    GitOps fans can take a spin on the free tier for pull-based deployment

    One-stop DevOps shop GitLab has announced version 15 of its platform, hot on the heels of pull-based GitOps turning up on the platform's free tier.

    Version 15.0 marks the arrival of GitLab's next major iteration and attention this time around has turned to visibility and observability – hardly surprising considering the acquisition of OpsTrace as 2021 drew to a close, as well as workflow automation, security and compliance.

    GitLab puts out monthly releases –  hitting 15.1 on June 22 –  and we spoke to the company's senior director of Product, Kenny Johnston, at the recent Kubecon EU event, about what will be added to version 15 as time goes by. During a chat with the company's senior director of Product, Kenny Johnston, at the recent Kubecon EU event, The Register was told that this was more where dollars were being invested into the product.

    Continue reading
  • To multicloud, or not: Former PayPal head engineer weighs in
    Not everyone needs it, but those who do need to consider 3 things, says Asim Razzaq

    The push is on to get every enterprise thinking they're missing out on the next big thing if they don't adopt a multicloud strategy.

    That shove in the multicloud direction appears to be working. More than 75 percent of businesses are now using multiple cloud providers, according to Gartner. That includes some big companies, like Boeing, which recently chose to spread its bets across AWS, Google Cloud and Azure as it continues to eliminate old legacy systems. 

    There are plenty of reasons to choose to go with multiple cloud providers, but Asim Razzaq, CEO and founder at cloud cost management company Yotascale, told The Register that choosing whether or not to invest in a multicloud architecture all comes down to three things: How many different compute needs a business has, budget, and the need for redundancy. 

    Continue reading

Biting the hand that feeds IT © 1998–2022