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.

Background

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

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

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

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." ®

Similar topics


Other stories you might like

  • Amazon delays return to office work until 2022 at the earliest

    Other Big Tech companies, however, still want workers in this autumn

    Amazon has delayed staff returning to its offices around the world from September this year to January 2022, as the Delta variant of the novel coronavirus continues to spread.

    “As we continue to closely watch local conditions related to COVID-19, we are adjusting our guidance for corporate employees in the U.S. and other countries where we had previously anticipated that employees would begin coming in regularly the week of Sept. 7,” the online bazaar said on Thursday. “We are now extending this date to Jan. 3, 2022. Our return-to-office timeline will vary globally in accordance with local conditions.”

    The pandemic has changed the way we work. Gone are the days where we need to commute into the office and work at our desks next to our colleagues. Recent surveys show that most people prefer working from home and don’t want to go back to the office much, if at all.

    Continue reading
  • All your DNS were belong to us: AWS and Google Cloud shut down spying vulnerability

    Security researchers found they could snoop on dynamic DNS traffic

    Until February this year, Amazon Route53's DNS service offered largely unappreciated network eavesdropping capabilities. And this undocumented spying option was also available at Google Cloud DNS and at least one other DNS-as-a-service provider.

    In a presentation earlier this week at the Black Hat USA 2021 security conference in Las Vegas, Nevada, Shir Tamari and Ami Luttwak from security firm Wiz, described how they found a DNS name server hijacking flaw that allowed them to spy on the dynamic DNS traffic of other customers.

    "We found a simple loophole that allowed us to intercept a portion of worldwide dynamic DNS traffic going through managed DNS providers like Amazon and Google," explained Tamari in a blog post. "Essentially, we 'wiretapped' the internal network traffic of 15,000 organizations (including Fortune 500 companies and government agencies) and millions of devices."

    Continue reading
  • Foxconn buys chip factory off Macronix in bid to break into the electric vehicle market

    Electronics giant must conquer its supply chain as US eyes domestic production

    Taiwanese electronics giant Foxconn has purchased a chip plant for $90.8m from its compatriot, Macronix International.

    "Macronix is pleased to see the subject 6-inch wafer fab continue to make its contribution to Taiwan as Foxconn commits to have the fab be used as an important base for Foxconn to reinforce its semiconductor development plan and to meet the demand of electric vehicles," said Miin Wu, chairman and CEO of Macronix, in a canned statement on Foxconn's website.

    The sales agreement includes Macronix's 6-inch wafer fab and equipment, but no employees, in Taiwan's Hsinchu Science Park and is planned to close by the end of 2021.

    Continue reading
  • THX Onyx: A do-it-all DAC for the travelling audiophile

    Hi-res, MQA, DSD, supports Apple Music's highest quality – but is it worth the hassle?

    Review Apple introduced hi-res lossless audio to its music service last month, but third-party hardware is required to enjoy it – if indeed the difference is audible. We took a look at the THX Onyx, a portable DAC and headphone amplifier that claims to be just the thing.

    There is a strange cocktail of ingredients that flavours the music and audio industry. There is a drive towards greater convenience, which means streaming music and true wireless, as popularised by Apple's Bluetooth-driven AirPods, first introduced in September 2016. Then there is a push towards higher quality, with vendors touting higher resolution such as 24-bit 192kHz digital, or exotic formats such as DSD (Direct Stream Digital), MQA (Master Quality Authenticated) – all of which are supported by the THX Onyx – and Dolby Atmos/Spatial audio, which is a new approach to surround sound.

    These two demands sometimes pull in opposite directions. Streaming audio has largely meant lossy compression, formats such as MP3 and AAC (Advanced Audio Coding), which reduce data size by omitting parts of the signal that are inaudible or hardly audible. Wireless has largely meant Bluetooth audio, for which none of the available codecs are lossless. Lossy compression at levels like Apple's 256 Kbps AAC is excellent and not an issue for most people yet there remains the nagging annoyance that it is potentially compromising quality for the sake of convenience and efficiency.

    Continue reading
  • Does the world need another cross-platform framework? Tough, here's JetBrains with Compose Multiplatform

    'A different way of thinking about applications' says project lead

    An open-source Kotlin framework for cross-platform applications, based on Jetpack Compose for Android, is now in preview.

    Google's Jetpack Compose is an official framework for building a user interface in an Android application, and reached version 1.0 last week, at the same time as the first stable release of Android Studio, 2020.3.1 or "Arctic Fox".

    Despite only just hitting 1.0, Google said: "There are already over 2,000 apps in the Play Store using Compose – in fact, the Play Store app itself uses Compose."

    Continue reading
  • Your Computer Is On Fire, but it will take much more than this book to put it out

    Detailed diagnosis of tech industry delusion falls short of prescribing a cure

    Book review Seasoned industry watchers will welcome Your Computer Is on Fire as a thorough and unflinching debunking of Big Tech's outlandish self-mythologising. They might even hope that governments, business, and the media organisations who buy into the barrage of propaganda start to ask a few important questions. But there are limits to this niche text that is at times prone to academic navel-gazing.

    In the 1990s, despite the outward differences between the industry big guns, the background hum was the same. The internet offered opportunity for all, ecommerce could lead to frictionless economics, software made people more productive, and companies more competitive. Such delusions survived the dotcom crash and financial crisis then re-emerged in the early days of social media as the Arab Spring became a use case for the positive impact of Twitter and Facebook. Together with that movement's difficult development, the nefarious exploitation of social media user data that contributed to the election of US presidential regime with ever-so-slightly insurrectionist tendencies should have given pause for thought.

    It's a wonder, then, that tech industry propaganda has barely shifted. Instead, it's a case of different tech, same tune. Last month, Google CEO Sundar Pichai told the BBC that AI would be the "most profound technology" that humanity will ever develop. Similarly, UK Cabinet Office minister Julia Lopez adopted industry language when she said that "now, more than ever, digital must be front and centre of government's priorities to meet user needs."

    Continue reading
  • Flushing roulette: Southern Water installing digital sewer monitors to prevent blockages

    Plan to deal with fatbergs NOT related to that £90m fine for dumping effluent into sea on England's south coast

    Where's there's muck there's brass, and there won't be many places more mucky than a sewer system as bidders for a network digitalisation contract in southern England are about to rediscover.

    According to a tender published this week, Southern Water is wading through the market to sniff out a supplier to "significantly and rapidly improve the visibility of the gravity wastewater network."

    "We plan to achieve this by installing 10,000's (up to 30,000 across Kent, East and West Sussex, Hampshire and the Isle of Wight) of sewer monitors and developing in parallel the associated analytics to make appropriate and effective use of the additional information to prevent sewer blockages developing into a pollution or flooding incident," the document states.

    Continue reading
  • Hey, AI software developers, you are taking Unicode into account, right ... right?

    Here's how to switch around account numbers, slip past moderation, and mix up names in production-level models

    Analysis Computer scientists have detailed ways in which AI language systems – including some in production – can be hoodwinked into making bad decisions by text containing unseen Unicode characters.

    Account numbers can be switched around, recipients of transactions changed, and comment moderation bypassed by special hidden characters, we're told. And it is claimed software built by Microsoft, Google, IBM, and Facebook can be potentially fooled by carefully crafted Unicode.

    The issue is that ambiguity or discrepancies can be introduced if the machine-learning software ignores certain invisible Unicode characters. What's seen on screen or printed out, for instance, won't match up with what the neural network saw and made a decision on. It may be possible abuse this lack of Unicode awareness for nefarious purposes.

    Continue reading
  • Q: Post-lockdown, where would I like to go? A: As far away from my own head as possible

    About 238,855 miles would do the trick

    Something for the Weekend, Sir? More good news for Team GB's Tokyo Games medal winners: you're going to the Moon.

    This is true because I read it. It was in a press release sent to me this week. "Olympic Medalists Get Free Ticket to the Moon," it says. That is going to be one expensive flight, with more than 50 medals awarded to Brits so far, and the Paralympics yet to begin.

    Unless… yep, here it comes in the first sentence. "LifeShip Inc. today announced they will be sending Olympic and Paralympic gold, silver, and bronze medalists' DNA to the Moon for free."

    Continue reading
  • Breaking Bad or just a bad breakpoint? That feeling when your predecessor is BASIC

    Here I am to save the day!

    On Call That Friday feeling is upon us again after a week of dealing with IT issues and dodging the gimlet gaze of the boss. Hopefully yours didn't involve some impromptu debugging in production. Welcome to On Call.

    Today are pleased to salute the return of Who, Me? contributor Susan, who previously regaled us with a tale from two decades ago.

    Susan's latest anecdote takes place in the months before Christmas 2001 when she found herself suddenly bereft of work. She had been enjoying the lucrative life of a Visual Basic 6 and SQL Server contractor before her employer of four years abruptly went bust, putting her dreams of a lavish festive holiday in the Scottish Highlands at risk.

    Continue reading
  • AI to be bigger than IaaS and PaaS combined by 2025

    $500bn a year to be spent on electro-brain and supporting tech vs $400bn on cloud infrastructure

    Analyst firm IDC has predicted that by the year 2025 more money will be spent on artificial intelligence software and services than on infrastructure-as-a-service and platforms-as-a-service.

    The firm on Wednesday published details of its Worldwide Semi-annual Artificial Intelligence Tracker, which predicted global spending of $341.8 billion this year – representing 15.2 per cent year-over-year growth.

    Growth will accelerate to 18.8 per cent in 2022, leading IDC to predict the market is on track to pass $500 billion annual spend by 2024.

    Continue reading

Biting the hand that feeds IT © 1998–2021