Nanotech - The Circuits Blog

Updated on May 6 at 6:35 p.m. PDT with additional comments from Intel.

A small brouhaha is erupting over Windows 7 and Intel processors. The hubbub is centered on which Intel processors will not support "XP mode" in Windows 7 and, by extension, which PCs will not support XP mode. Retail laptops may be one of the most prominent segments affected.

Sony Vaio laptops sold at retail stores are among a number of models from a variety of PC makers that have processors that don't support Windows 7 XP mode

(Credit: Best Buy)

What is XP Mode? Here's how Ina Fried of CNET News describes it: "XP mode consists of two things, the Windows Virtual PC engine and a licensed copy of Windows XP Service Pack 3 as a packaged virtual machine. Although neither piece will be included in the Windows 7 box, XP Mode will be a free download for those who have a license to Windows 7 Professional, Windows 7 Enterprise, or Windows 7 Ultimate."

XP Mode (XPM) is aimed at businesses that have Windows XP-specific applications that they need to run on Windows 7. XPM allows XP applications to run seamlessly on Windows 7, according to Microsoft. The catch: Intel processors must have Virtualization Technology, or "Intel VT," in order to run XPM. (I won't cover Advanced Micro Devices processors here but will address AMD in a later post.)

Ed Bott's Microsoft Report says that "some of the most popular PCs on the market today...won't be able to use the vaunted Windows XP mode in Windows 7."

Bott lists Intel desktop and mobile processors that will and will not support XP Mode here and here, respectively.

Intel mobile processors may be the most problematic in supporting XP mode; not because of the raw numbers--most newer Intel mobile processors do, in fact, support Intel Virtualization Technology--but because a disproportionate number of those that do not have VT (and therefore don't support XP mode) are laptops sold at retail. (And, undoubtedly, some small businesses purchase laptops at retail.)

In the Core 2 mobile camp, the P7350/7450, the T5200/5250/5270/5300/5450/5470 series, and the T6400/6570 do not support VT, according to Bott's blog. And this can be confirmed on Intel's Web site.

A quick glance at Best Buy shows a somewhat lengthy list of laptop SKUs (models) with, for instance, the T6400 (non-VT) processor. The list includes Dell Studio, Toshiba Satellite, HP Pavilion, Sony Vaio, Asus, and Gateway laptops.

In the $600 to $899 laptop range, I found about 30 different SKUs with T6400 processors, though it should be noted that some of these SKUs are simply models with slightly different configurations.

And a quick search on CNET Shopper turns up a number of consumer models with the T5270. The point? To state the obvious, consumers will have to verify which processor their laptop has.

In an Intel blog, Nick Knupffer asserts this won't be a big issue. "Having VT on these consumer laptops is not going to be an issue--because the consumer versions of Windows 7 (Starter, Home Basic, and Home Premium) do not include Windows XP Mode," he writes.

And Intel, in a statement, had this to say. "Intel introduced its Virtualization Technology in 2005 and has shipped over 100 Million chips with the feature. Windows XP Mode is targeted for business customers. It is available on the mid to higher end versions of Windows 7 and is supported in hardware by many Intel processors."

Intel continued: "Intel vPro technology PCs are required to have an Intel VT capable CPU and Intel VT capable BIOS. They are the best platforms for testing and deploying Microsoft Windows Virtual PC and Windows XP Mode."

Some researchers claim that Intel has a serious chip bug on its hands. But that all depends.

Security experts who are into the arcana of chip security may find "CPU cache poisoning" riveting and serious stuff. Others, however, may simply scratch their heads and move on.

But let's not move on too quickly. First, a quote from an abstract of the paper (PDF) that has some of the chip world abuzz. "In this paper we have described practical exploitation of the CPU cache poisoning...This is the third attack on SMM (system management mode) memory our team has found within the last 10 months, affecting Intel-based systems. It seems that the current state of firmware security, even in case of such reputable vendors as Intel, is quite unsatisfying."

Joanna Rutkowska, who exposed the potential of the so-called Blue Pill flaw in August 2006 and who founded Invisible Things Lab, wrote that excerpt (along with colleague Rafal Wojtczuk) and obviously takes this very seriously.

As do others. Not worried yet? "This is the scariest, stealthiest, and most dangerous exploit I've seen come around since the legendary Blue Pill!," writes Jamey Heary in a Network World blog. He is a consulting systems engineer for Cisco Systems.

So now that we know it's scary, what could happen in a worst-case scenario? Suffice to say that gaining access to "privileged" SMM memory would essentially allow hackers to do anything to the target PC that they want. The question is, would they actually take advantage of this particular opening?

"If a hacker can use this new exploit to embed a SMM rootkit (malware) they would have ultimate control over the box (computer). Additionally, it would be virtually undetectable," Heary wrote in response to an e-mail query. But he also added: "In a nutshell. This exploit is very serious and needs to fixed. But...I don't see a mass virus or worm using this. The attacks will be targeted. A rootkit must be perfectly matched to the hardware. This makes mass infection more difficult."

Rutkowska and Wojtczuk, in the abstract, say that the paper discusses "how to practically exploit this problem, showing working proof of concept codes that allow for arbitrary SMM code execution. This allows for various kind of abuses of the super-privileged SMM mode, e.g. via SMM rootkits."

Who can do this? "We assume that the attacker has (what is in practice)...equivalent to administrator privileges on the target system, and on some systems, e.g. Windows, also the ability to load and execute arbitrary kernel code," write Rutkowska and Wojtczuk.

And what systems are potentially vulnerable? Though both Intel and Rutkowska say the "attack" presented in the paper has been fixed on some systems, Rutkowska goes on to say: "We have however found out that even the relatively new boards, e.g. Intel DQ35 are still vulnerable (the very recent Intel DQ45 doesn't seem to be vulnerable though). The exploit attached is for DQ35 board--the offsets would have to be changed to work on other boards (please do not ask how to do this)." (Here is a list of Intel motherboards she refers to.)

These motherboards are used with Core 2 Quad, Core 2 Duo, Pentium, and Celeron processors, according to Intel's Web site.

Intel has addressed the matter this way: "We are working with these researchers. We take this research and all reports seriously. Currently as far as we know, there are no known exploits in the wild," Intel spokesman George Alfs said in a written statement.

One point worth noting is that this is not an Intel errata per se, which Intel typically details in processor specification updates. This is a theoretical attack from a malicious hacker. Nevertheless, users can minimize the risk by keeping up-to-date on patches and on operating system and security suite updates. Particularly important are BIOS (basic input/output system) and firmware updates for the processors and motherboards referenced above.

So, what is the average user to make of all of this? Security attacks and security vulnerabilities have been around since (computer) time immemorial (in the relatively brief history of mass-market computing). A report from U.K.-based technology Web site The Register in 2006, for example, suggested that people should not purchase Core 2 Duo systems--now widespread worldwide--because of security vulnerabilities and cited an open-source expert, who prophesied doom and gloom for the Core 2 Duo architecture.

Then there's the whopper of them all--and a flaw very different in nature from the SMM vulnerability discussed above--the show-stopping 1994 Intel FDIV bug, discovered by Professor Thomas Nicely, then at Lynchburg College in Virginia. Also referred to as the floating-point bug, it wasn't a flaw exploitable by malicious hackers; rather, it was a bug in Intel's original Pentium floating-point unit. Certain arcane floating-point division operations done on these processors would generate incorrect results.

This bug, covered prominently by The New York Times and CNN at the time, actually had virtually no affect on users, except causing them to panic and, as a consequence, some insisted that Intel provide them with new processors. The recall cost Intel close to a half-billion dollars.

Updated on January 19 at 8:15 a.m. PST with additional information throughout.

On Sunday, Intel instituted broad price cuts on processors, spanning the Core 2 Quad, Core 2 Duo, Pentium dual-core, Celeron, and Xeon product lines.

Some of the cuts are in response to Advanced Micro Devices' recently-introduced Phenom II "Dragon" desktop platform. AMD's Phenom II X4 940 (3.0GHz), for instance, is priced at $275.

Intel cuts were concentrated on quad-core chips like the Q9650 (3.00GHz), reduced 40 percent, to $316 from $530, to counter AMD's Phenom II. But Celeron processors received some of the largest reductions. The mobile Celeron 570 (2.00GHz), was slashed 48 percent, to $70 from $134, for example.

Some Xeon processors also received hefty cuts. The price on the X3370 (3.00GHz), for instance, was cut 40 percent to $316 from $530.

The Pentium dual-core E5200 was reduced 24 percent to $64 from $84.

The mobile Core 2 Duo P8600 was shaved 13 percent, to $209 from $241. The P series of mobile Core 2 Duo processors have a thermal envelope (Thermal Design Power or TDP) of 25 watts versus 35 watts for mainstream mobile chips.

Intel also introduced new processor models such as the desktop Core 2 Duo E7500 (2.93GHz), priced at $133 and the Core 2 Quad Q9550s (2.83GHz), which is a low-power variant--with a thermal envelope of 65 watts--of the 95-watt Q9550.

Update at 9:00 a.m. PDT July 9 adding new information that links Larrabee cores to Pentium technology

The "rumor" is true: Intel's future Larrabee graphics chip does use processing cores based on the company's venerable Pentium chip.

Intel currently offers a dual-core Pentium processor; Intel will use a core based on older Pentium technology in Larrabee.

(Credit: Intel)

Larrabee, due in 2009-2010, is a high-end graphics chip with many processing cores targeting market segments that Nvidia and AMD-ATI now dominate.

Here's the starting point for the Larrabee-is-a-Pentium theme. The Web site Custom PC cited an article from German-language Heise, in which Custom PC says--via a link to a Babel Fish translation of the Heise article--Intel Senior Vice President Pat Gelsinger discusses Larrabee and Pentium.

Though Intel is refusing to confirm the report, the Heise article is true, according to a source who has accurate data about Larrabee.

The cores in Larrabee will have a "basic commonality" with the P54C Pentium, the source said. The P54C was introduced back in 1994.

The technology Web site, ARS Technica, is also saying that Larrabee is based on the P54C Pentium core.

How many of these small cores (which Intel calls mini-cores) will Larrabee have? Heise says that Larrabee will have 32 processing cores.

Intel has been casting doubt on the reports in an attempt, apparently, to keep the information under wraps. Intel said this Tuesday: The Heise "story does not say that Pat (Gelsinger) confirmed anything. Neither number of cores, not the type of cores. The Babel Fish translation is misleading."

Intel is expected to provide further details on Larrabee at upcoming conferences including Siggraph next month in Los Angeles.

Larrabee has been ridiculed by Nvidia CEO Jen-Hsun Huang as nothing more than a "PowerPoint slide" since few tangible particulars have yet to emerge about the processor though Intel mentions the future chip often.

Slashdot link here: Larrabee Based On a Bundle of Old Pentium Chips.