Linus Torvalds: Intel can never again be trusted, under any circumstances. The future is ARM64.

Linus Torvalds, Santa Fe, New Mexico, 2016. [Source: Faces of Open Source]

by V Lexander [07 January 2017]

There are a very few individuals out there who know what the hell they’re talking about and who actually understand technology, and in particular, exactly how the Internet functions. Two of those individuals are the two men who invented the Internet, Vint Cert and Robert Kahn. Another one is Linus Torvalds.

As with Cerf and Kahn, the position and importance of Torvalds cannot be understated. Without his invention of the Linux kernel — the development of which he continues to direct and oversee — the Internet as we know and use it today would not exist. The Linux kernel, which when combined with a basic set of tools from Richard Stallman‘s GNU Project, forms a complete and absolutely free operating system known as GNU/Linux, which powers the majority of the servers connected to the global Internet as well as domestic internets (the term “internet,” as used in the lowercase, refers to internetworks in general and not the global Internet — they are not the same, as both Cerf and Kahn have clarified numerous times, while continually ignored and even insulted by mass media outlets) and organizational intranets.

But the development of GNU/Linux was only the very beginning of an open source revolution that led to the Linux kernel being used as the base for a number of highly significant operating systems, the most important of which is Android. Any smartphone, mobile or embedded device that uses the Android operating system (which is not GNU/Linux and has little to no connection to GNU applications and projects) has Linux at its core, which is exactly what a “kernel” is, the lowest level core of an operating system that makes it possible to use a device in the first place.

The reason why Linux made it possible for the Internet to undergo such rapid and unprecedented growth has been due to the fact that it was released under a free and open source license (and continues to be), unlike proprietary operating systems such as Microsoft Windows, macOS (formerly Mac OS X and previous to that, MacOS/Macintosh System) and the various commercial UNIX systems.) Ian Murdock, the second most important individual in Linux, developed the first fully complete version of GNU/Linux in 1993 called Debian, which allowed commercial Internet Service Providers (ISPs), very few of which could have even afforded the hopelessly Byzantine licensing registrations and fees associated with Microsoft and UNIX vendors. Alongside the rapid development and deployment of Debian in 1993, a visionary startup in North Carolina called Red Hat released the first commercial enterprise variant of GNU/Linux to provide support for proprietary drivers (which was necessary at the time in order to use GNU/Linux with most graphics and sound hardware), as well as full technical support and on-site training and deployment at extremely low cost, which helped drive Fortune 500 corporations away from Microsoft and UNIX.

This is how the Internet has evolved into the all-encompassing and near infinitely scalable Matrix that it is today, which will eventually be superseded by bleeding edge next generation technologies being developed at Internet2. And it all began in 1990 with Linus Torvalds, a low-income university student who had received an Intel 386 desktop system as a gift, at the time an expensive workstation that most people wouldn’t have been able to afford. This was also in the same year that Tim Berners-Lee invented the World Wide Web, the technical name of which is the Hypertext Transfer Protocol.

To say that Torvalds knows and understands Intel microprocessor technology is an understatement. All of this is very well known and not in dispute by anyone who knows anything about technology.

What is not so well known is that most, if not all of Intel’s microprocessor research and development has been conducted at its facilities in Israel, a country that has long since become one of the global epicenters of bleeding edge research. This has been a major factor in keeping Israel’s brain drain in check, which is endemic and potentially threatens its position as a leading innovator in science and technology, not just in computing but also in medicine and bioengineering.

Such a fact, in and of itself, would not be much of an issue were it not due to the geopolitical conflict between Israel and Iran. After the disastrous blowback suffered by Israel in its failed attempt to derail and ultimately destroy the Iranian civilian nuclear program via Stuxnet, Iran initiated an unprecedented overhaul of its domestic science and technology infrastructure, beginning with the full elimination of Microsoft Windows across the board and building a stronger and now virtually impenetrable firewall at all its research and nuclear facilities, all due to flaws exploited by Stuxnet and other US and Israeli sponsored Internet warfare efforts. Both the CIA and NSA themselves criticized Stuxnet and the way it was deployed, predicting exactly what has since transpired. While both US agencies were involved in its development, the unleashing of Stuxnet was more or less done preemptively on the part of the Israeli military-industrial complex. Similar efforts (mainly on the part of the US) against Russia and China at the time also led to similar failures, as such exploits were dependent upon the success of Stuxnet.

In light of such a catastrophe, how then could the US, Israel and its Arab allies (Saudi Arabia and the United Arab Emirates, in particular) undermine the Iranian nuclear program? To make matters worse, in the years since Stuxnet, both Russia and China have invested billions into radical overhauls of their own military-industrial complexes to become fully independent of foreign technology. Russian, Chinese and Iranian computer scientists all agreed upon an inevitable reality: that it wasn’t enough to simply stop using Microsoft Windows and off-the-shelf equipment developed by US and Israeli firms. They needed to build entirely new microprocessor architectures from scratch, because Intel and AMD could no longer be trusted by these countries. From their perspective, they weren’t wrong. But this would be a daunting task and would take years to complete. It’s one thing for these countries to develop cutting edge microprocessors for their supercomputer arrays, missile defense systems and nuclear facilities, but affordable processors for the consumer desktop and mobile device market? This wasn’t going to be happening anytime soon. Even Russian, Chinese and Iranian scientific and nuclear research facilities would remain long dependent upon Intel processors until stability issues with their own domestically manufactured processors could be fully resolved.

For both the US and Israel, these facts led to a new and diabolically ingenious solution. Many US and Israeli engineers responsible for Stuxnet and Internet warfare efforts were critical of Stuxnet due to the impact of the Morris worm of the 1980s. Exploits and malware launched on a global scale are far too prone to numerous variables that are impossible even for teams of scientists and engineers to take into consideration and circumvent. The results of Stuxnet proved every single one of their criticisms beyond all reasonable doubt.

The better and more effective solution was to develop a Trojan horse at the lowest hardware level possible, and that would be the microprocessor itself. These microprocessor-level exploits (incorrectly referred to as “flaws,” as these were deliberately designed and implemented) have since been revealed to be known as “Meltdown” and “Spectre.” These are not media buzzwords that some journalist came up with. These are internal code words unofficially used within Intel to refer to these exploits. The fact that these exploits cannot be corrected at the microcode level and require developers to circumvent the vulnerabilities at the operating system level prove without a doubt that Intel deliberately engineered and implemented these exploits. As Torvalds explains in his email below, they have knowingly been selling deliberately compromised processors and royally screwing over their consumers everywhere.

What he doesn’t explain nor acknowledge — and he wouldn’t be able to, even if he wanted, due to political sensitivities and current geopolitical reality — is why Intel did this. When you put two and two together, it isn’t all that difficult to figure out. With Russia, China and Iran all moving away from Intel and any and all proprietary hardware technologies (most of which have been designed and developed in the US and Israel) and developing their own hardware based either on free and open source standards or their own domestic proprietary designs, “Meltdown” and “Spectre” were intended to be used in potential efforts to attempt damaging or even outright destroying enemy military-industrial complexes, chiefly those of Russia, China and Iran. As the saying goes, all is fair in love and war.

But in doing so, Intel has totally and completely destroyed their credibility in a way that no one had ever thought possible, and anyone who thinks Intel engineers are incompetent and/or stupid are either hopelessly naive or just plain ignorant. AMD, with its close ties to the Saudi and Emirati regimes, as well with the Israeli military-industrial complex, has found itself in an even more precarious and catastrophic position.

The long-term solution to all of this is are microprocessor architectures that are fully free and open source. We’re not there yet and won’t be for many years. Until that happens, Torvalds suggests that ARM64 may be a potential answer to these problems. It remains a proprietary architecture, but one that is sufficiently open due to its unique licensing structure that it cannot be subject to such insidious and inherently evil schemes that corporations such as Intel and AMD are able to so easily and covertly impose on us all.

But this is all subtext, because it’s not just a potential or possible solution. It is the solution. There can be no doubt: ARM64 and server physicalization are the wave of the future and it’s been happening for some time now. All Intel and AMD have done is accelerate the inevitable, just as Microsoft, Apple and the major UNIX vendors did when faced with the existential threat of GNU/Linux.

We live in interesting times indeed. All we have to do is keep following that white rabbit toward our ultimate liberation from the tyranny of technological dystopia.


 

From Linus Torvalds <>
Date Wed, 3 Jan 2018 15:51:35 -0800
Subject Re: Avoid speculative indirect calls in kernel
On Wed, Jan 3, 2018 at 3:09 PM, Andi Kleen <[email protected]> wrote:
> This is a fix for Variant 2 in
> https://googleprojectzero.blogspot.com/2018/01/reading-privileged-memory-with-side.html
>
> Any speculative indirect calls in the kernel can be tricked
> to execute any kernel code, which may allow side channel
> attacks that can leak arbitrary kernel data.

Why is this all done without any configuration options?

A *competent* CPU engineer would fix this by making sure speculation
doesn't happen across protection domains. Maybe even a L1 I$ that is
keyed by CPL.

I think somebody inside of Intel needs to really take a long hard look
at their CPU's, and actually admit that they have issues instead of
writing PR blurbs that say that everything works as designed.

.. and that really means that all these mitigation patches should be
written with "not all CPU's are crap" in mind.

Or is Intel basically saying "we are committed to selling you shit
forever and ever, and never fixing anything"?

Because if that's the case, maybe we should start looking towards the
ARM64 people more.

Please talk to management. Because I really see exactly two possibibilities:

 - Intel never intends to fix anything

OR

 - these workarounds should have a way to disable them.

Which of the two is it?

                   Linus


[Source: https://lkml.org/lkml/2018/1/3/797]

Leave a Reply

Your email address will not be published. Required fields are marked *