We interrupt our regularly scheduled program to bring you breaking news... Intel has acquired WindRiver Systems for $884 Million. I can certainly see where this makes sense for Intel. They've been pushing in the embedded space for quite some time with the Atom. And, WRS's offerings play well into an apparent Intel strategy for moving further into the Mil/Aero arena. The question that remains is what will this move do for Cavium, Raza, FreeScale, etc. that had been using WRS for their Linux partner?

Does this drive those companies to folks like MontaVista? Does this make MontaVista a takeover target for FreeScale or even IBM? Things that make you go hmm.... Clearly, the impact of this move will have reverberations into the market for quite some time. And, who will Green Hills have to bash now that WRS has been absorbed into the body? Only time will tell...

The link for the news is at: http://www.windriver.com/news/press/pr.html?ID=6921

Enjoy.

With the economic downturn and the imminent demise of Windows Vista, more folks are looking seriously at Linux as a development platform for embedded systems. However, most of us live in a Micro$oft-centric world. We have Exchange servers, Outlook email, Word documents, Powerpoint presentations and a management infrastructure that only understands MS Windows. How can the developer who wants more tools and a more stable development environment survive in the typical corporate world?

Over the next several weeks, I'll be outlining the process of tooling up in Linux while not losing any of the key connections to your management infrastructure. This ranges from the hardware needed to run Linux, to the infrastructure connections to devices like printers and wireless, to the software needed to keep you in contact with the rest of your peers. I'll be outlining several alternatives along the way as well. So, if one approach isn't to your liking, you can try one of the others.

Before we get started, let me tell you the good and bad of Linux. The good is that there is never only one way to do anything in Linux. In almost 10 years of working with Linux, I've never encountered a roadblock so severe that there simply wasn't a way around it. Now the bad part: there is never only one way to do anything in Linux. There are usually at least 6 different ways to do the same thing in Linux. This means that you'll have to pick one approach and go with it until either you get things working, or you find that the technique won't work for you and you have to choose a different route.

That being said, I can assure you that you can live quite comfortably in Linux and still have complete connectivity with your peers. You can even skin the UI in Linux to make it look like Windows XP, Vista, or even OS/X. So, you won't lose the look and feel that you've become comfortable with and the IT wonks can't waltz by your desk and tell that you're not using the official O/S.

So, stay tuned to this channel as I outline what it takes to do embedded development in Linux...

I had a friend of mine a few months back tell me that Intel's Atom processor was a "game changer" for embedded systems. The theory goes that Intel is now in the low-power marketplace and the days of processors like the ARM and Power cores are numbered. After I stopped laughing, I took some time to really think it through from his perspective.

He is in the consumer electronics biz. He makes DVRs. For devices that are always plugged in and generally always on, I suppose that the Atom could be a significant player. Previously, the most power efficient x86 variants like the P8400, were in the 25 watt range. The Atom drops that down to about 2 watts. That's very good for the processor. But, the Atom's southbridge like the ICH7 draws close to 11 watts by itself. So, this means that a current technology Atom design will be in the 10-15 watt range for just the processor and southbridge.

This is laugable when compared to an OMAP 3430 at less than 500 miliwatts. But, Intel is working on integrating the southbridge and producing a single SOC. That should bring the power requirements down even further. We'll have to wait and see what Intels latest deal with the SOC manufacturer TSMC reveals.

So, his point was well taken. For a number of platforms like DVRs, set-top boxes, TV sets, etc. having an x86-compatible device at even half of the power requirements of a typical laptop has some advantages. First of all, fanless operation becomes a real possibility. And, you can eliminate a lot of the hassles of cross development. Just use a standard x86 as the development platform and run native code.

However, from a "green" perspective, leaving even the Atom running 24/7 is not very efficient. There are already way too many devices out there where "off" is not really off. If we want the flexibility of instant on, then we have to be willing to pay the price of the parasitic power draw or come up with even better power management capabilities. Intel has the engineering talent to make something like this happen, but the legacy of the 8086 (introduced in 1978) may be working against them. Time will tell.

So, from the viewpoint of a battery-operated device such as a cell phone or handheld game device, I don't see the Intel Atom as much of a player -- yet. For devices with AC power available all of the time, yeah I think maybe the Atom is a game changer. It can certainly cut development costs and Intel is pricing these things aggressively. Aggressively enough to make it cheaper to use the 1.6 GHz Atom instead of that 300 MHz 486-based SOC I've been using for some industrial applications. BTW, I apologized for laughing...

Hmm... OK, it's been almost three years since Intel sold the XSCALE line to Marvell. Back just after the sale, there was a lot of noise about the Monahans part (PXA-320) and how power efficient it was, etc. We all waited, but the PXA-320 never really made the splash or the inroads into the cell phone market that it was supposed to.

In the mean time, TI has come out with multiple OMAP variants that are making news. One only needs to look at the BeagleBoard or the Gumstix Overo to see that OMAP is doing well. FreeScale has also made good progess with the i.Mx series. Good low-power processors that have a reasonable support infrastructure. So what happened to XSCALE?

My theory is that XSCALE suffers from Marvell's paranoia. The same paranoia that caused so many people to avoid Broadcom parts like the plauge. It's just too difficult to get information, data sheets etc. out of Marvell. Every time I've dealt with Marvell, it's been like pulling teeth to get any information out of them. I've got to get V.P. level signatures for NDAs that only last for 30 days so I can download data sheets with so many erratta that I have to keep re-upping the NDAs just to stay on top of it.

As a small unit production developer, I feel like Marvell is afraid that I'll steal their proprietary IP and produce a copy of their chip if they tell me what registers exist so I can port and O/S or a driver to their silicon. I just want to use the silicon. I need information to make that happen. If it's difficult to get the information, I'll use someone else's silicon who will provide me the info like TI or FreeScale.

OK, I've heard that for huge volumes, Marvell will provide great service. I don't doubt that for million unit quantities they will assign a dedicated resource to me. But, the majority of business volume in the U.S. is done by small businesses. And, if it's difficult to get data sheets as a small business, engineers who move to large business will still avoid the parts because of past experience.

So, I guess my point is that I'm not really seeing much buzz about XSCALE in the past couple of years since Marvell took it over. It's certainly not because it wasn't a good architecture. Look at all of the PXA-270 based phones that where produced when Intel ran the line. I believe that XSCALE is dying because of paranoia about data sheets falling into the hands of the open-source community.

My therory is that if Marvell wants XSCALE to stage a comeback, they need to change their data access policies. They need to embrace open-source and create a low-cost platform for the open-source developers to be able to port and experiment with. Without cooperation with the community, XSCALE will become a backwater eddy in the processor world and eventually die out due to lack of interest. I'm going to play with my BeagleBoard now...