Android on x86 and Binary Translation

So the other major and obvious piece of the puzzle is what changes were required to make Android and all of its applications run on x86. Android itself has already been built for x86 before, and again we’ve seen and played with it running all the way back at IDF 2010. That part of the puzzle is relatively understood, but the devil again is in the edge cases. Among the things that need massaging for x86 are the Dalvik VM, x86 JIT, NDK, and JavaScript engine.

Android itself is actually an ideal platform for Intel to target, as the vast majority of Android applications in the Play Store run atop the Dalvik VM and use the Android Framework (75–80% are commonly cited numbers, depending on how you’re counting). The rest either are Dalvik VM applications that run and use JNI (Java Native Interface) libraries that are built for ARM only, or NDK (Native Development Kit) applications. So where does Intel’s binary translation secret sauce fit into all this? Simply put, Intel’s binary translation is the mitigation for both libraries and NDK applications that haven’t yet been ported to x86, and allows the device to expose itself as supporting two application binary interfaces (ABIs), both x86 and ARMv5, in fact this is easy enough to see upon superficial inspection of build.prop:

ro.product.cpu.abi=x86
...
ro.product.cpu.abi2=armeabi

In the case of Dalvik applications, developers don’t need to do anything. Thankfully again this is the vast majority of Android applications you encounter on a daily basis - they just work, given that Intel has made Dalvik work with x86 and spit out the right machine code.

NDK applications are also easy enough to mitigate - the developer simply needs to recompile the NDK project, which supports ARMv5 (‘armeabi’), ARMv7 (‘armeabi-v7a’), and x86 (‘x86’). Building for x86 will deliver code that’s tailored (unsurprisingly) exactly to the Saltwell CPU feature set, or more explicitly what you’d get by running GCC with the compiler flags “-march=i686 -msse3 -mstackrealign -mfpmath=sse” - this is all outlined in the CPU-ARCH-ABIS.html document as part of the NDK documentation. The resulting APK can be packaged as a “fat binary” with machine code for all three platforms, and upon install only the proper one is unpacked and installed.

The remaining two cases are where binary translation come in. In the case of applications that haven’t been rebuilt with the NDK to target x86, the binary translator magic kicks in and translates the armeabi version into x86. The same applies for applications that request some JNI libraries that are currently ARM only.

Intel outlines this in a number of slides which have made their way online, and the process is virtually completely transparent to end users and Dalvik applications. The x86 compatible Dalvik VM is a part of the OS, as are the ARM to Atom BT phase for JNI libraries. ARM native NDK apps on the other hand are translated by Intel in the cloud, validated against Intel's Android x86 emulator and pushed to the Play Store. The point is the bulk of binary translation happens away from the device itself and running on much faster Xeons in the cloud. As binary translation requires more cycles than natively running the code, which in turn consumes additional power, this was the only route for Intel to ensure that Atom would remain power efficient (and high performance) even on non-native NDK apps. Update: Intel has clarified and informed us there is no cloud aspect to binary translation, it is 100% done on the device for ARM NDK applications.

It's still unclear just how long this process takes after a developer has uploaded a non-x86 NDK app to the Play Store, or what happens if the process fails to validate for whatever reason (Does Intel get in touch with the developer? Is the app forever excluded?). Intel is being unusually vague about how all of this works unfortunately.

The combination of all of these efforts should result in over 90% of the apps in the Play Store working right away. What about in our experience? We discuss that next.

Software: Nearly Flawless

The X900 that I was sent came running Android 2.3.7, which is the latest version on the 2.3 branch. Xolo intends to deliver an Android 4.0.3 update later, and Intel internally has its own 4.0.x image stable and ready to go, which we’ve seen running on the FFRD a bunch before. It’s a bit odd to see things going this way when 4.0.x is clearly already ready, but no doubt some logistical issues with carrier support are the final hurdle. I’m eager to check out Intel’s 4.0.x port and intend to update when that happens.

 

Xolo and Intel have basically left things entirely stock with the X900. The notifications shade has one minor positive change - inclusion of the power controls, and Swype is bundled in addition to the stock Android 2.3 keyboard. There’s one Xolo Care support application preloaded, and basically nothing else. I can honestly say this is the least preloaded junk I’ve ever seen on a non-Nexus device.

 

So the next logical step is talking about how well Android and its apps work on x86 in practice, and the answer is unsurprisingly that almost everything is perfect. I installed about 80% of all the Android applications I’ve ever installed on any Android phone (thanks to the new Google Play ‘All’ tab) and nearly all of them worked perfectly. In fact, all of my daily driver applications work flawlessly: Twitter for Android, Baconreader, Speedtest.net, Barcode Scanner, Astro, Dropbox, Facebook, GPS Test Plus, GPS Status, Instagram, IP Cam Viewer, GTA III, Remote Desktop, Swiftkey X, and WiFi Analyzer all work perfectly.

 

That said there are indeed a few edge cases where things don’t seem to be perfect. For one, Flash 11 isn’t available for the X900, and throws an error in the market. The device does come preloaded running Flash 10.3 however, which gets the job done although is a bit dated. In addition, although Netflix would download, the installer would throw a ‘package file invalid’ error upon install. This is what leads me to think there’s some APK interception in the cloud and perhaps translation up there, and Netflix DRM not translating, but that’s speculation. Other than this, everything else I encountered works flawlessly, I wager your average Android user wouldn't be able to tell that this is running on a completely different architecture.

Medfield: Intel in a Smartphone Performance
Comments Locked

106 Comments

View All Comments

  • vol7ron - Wednesday, April 25, 2012 - link

    I doubt windows would expect PCI channels lol. Though, it might need drivers to operate.

    Everything you need is on the phone for windows to operate (Screen, CPU, Video, RAM and Disk space) exists, even though Windows doesn't require it all. Though, Windows does need some way to communicate with those devices (device drivers), which Win7/etc probably doesn't have.
  • Shadowmaster625 - Wednesday, April 25, 2012 - link

    A few years from now it is likely I might be able to acquire one of these for dirt cheap. (Broken screen, etc) I would use it just for an ultra low power ultra low profile *single-function* pc. I would very much like to know if this hardware can run windows 7. It doesnt need to run well, it just needs to be able to go on the web and do basic things similar to an atom nettop.
  • superPC - Wednesday, April 25, 2012 - link

    it won't run windows 7. unlike windows 8, windows 7 requires standard RAM not LPDDR. windows 7 also requires some form of PCI.
  • Musafir_86 - Wednesday, April 25, 2012 - link

    -Excuse me, but IMHO, the type of physical RAM shouldn't matter. If not, we couldn't be able to load these OSes on VMs at all. :)

    Regards.
  • B3an - Thursday, April 26, 2012 - link

    Why would you even want to run Win 7 on this when Win 8 would clearly be WAY better suited, not to mention it also uses less resources and RAM while remaining faster/snappier than 7.
  • rahvin - Friday, April 27, 2012 - link

    I'd imagine he wants to know because Windows 8 is going to be only slightly less successful than Vista. Personally I'd guess around 5% of the Vista sales. It's a disaster in waiting unless they make dramatic last minute changes. You should try using it.
  • joshv - Wednesday, April 25, 2012 - link

    I am not sure why this chipset matters. Intel usually wins on x86 compatibility with older software. In the phone space there is no existing x86 code, and in fact they are stuck emulating another ABI - so they will be slower and less efficient that competitors that implement that ABI natively.

    That leaves Intel to compete on price/performance alone in a market where their competitors have 99.9% of the market. An odd position for Intel.

    Perhaps this makes more sense in a Windows 8 tablet?
  • Impulses - Wednesday, April 25, 2012 - link

    Its netbooks all over again, on a much bigger scale. ARM is moving upscale, if Intel doesn't start competing directly they will eventually start ceding some existing market share (when tablets/laptops start to overlap more, and the writing's on the wall with Windows for ARM).

    Only difference is they're up against a capable rival(s) as opposed to a limping AMD, so they can't just come out of the gate strong and them dog it and let the lower end market stagnate in order to maintain profits.

    This is a small first step but it'll allow them to ink more deals and possibly cement a strong foundation for upcoming Win8 ARM tablets which is probably their bigger long term concern.
  • dcollins - Wednesday, April 25, 2012 - link

    Did you even read the article?

    The x86 vs ARM issue is mostly a non-issue that will be completely resolved within a few months. Dalvik apps are JIT compiled to ARM and x86 and will perform similarly. In fact Intel might have an advantage here because they have the best compiler engineers in the world with decades of experience in generating high performance x86 code. NDK apps will generally be supported natively; developers only have to check a box to include x86 binaries. Even Apps that aren't compiled with x86 support are translated prior to installation on the users device. Nothing about the instruction set makes Medfield slower than ARM.

    Performance today is comparable to modern ARM processors even when running an out of date, slower OS. Performance in 4.0.x should match or outperform even Krait. Graphics performance is middle of the road, but that's a major concern for many smartphone buyers (myself included). Even in benchmarks that purposefully stress mutliple cores, Medfield holds its own against the many cored competitors. Real world usage is more lightly threaded.

    Browser performance is the most important metric for my usage and here Intel performs extremely well. If Medfield is available in a 4.x phone when it comes time for me to upgrade, I will seriously consider it versus a Krait based offering. Now imagine a next generation Atom build on 22nm with dual core, hyperthreading and possibly OoO execution: that chip will eat A15s for breakfast.
  • dcollins - Wednesday, April 25, 2012 - link

    edit: "Graphics performances... is NOT a major concern"

Log in

Don't have an account? Sign up now