Juno dev platform lays foundation for 64-bit Android on ARM

The folks at ARM have a bit of a chicken-and-egg problem that any creator of a new CPU instruction set is likely to face. How exactly do they enable the creation of ARMv8-compatible chips, devices, drivers, and software without any functioning hardware available for testing? More acutely, how do they make sure the transition to the 64-bit ARMv8 instruction set happens quickly enough to keep up with the crazy-short design cycles of today’s tech gear?

The answer, it turns out, is for ARM to build its very own reference platform, complete with a six-core custom SoC. That platform is dubbed Juno, and ARM says it’s just now becoming available to customers who could make use of that sort of thing—think SoC makers, kernel and hypervisor developers, the creators of driver software, and the low-level programmers who build things like game engines.

The news of an ARMv8 reference platform’s first availability in mid-2014 may seem surprising given the tremendous momentum ARM’s 64-bit ISA already appears to have in multiple markets. After all, Apple has been shipping 64-bit hardware for some time now, and Google announced the Developer Preview of the 64-bit Android L release just last week at its I/O conference. Still, until Juno came along, the only ARMv8-compatible hardware out there was Apple’s custom SoCs, which are closed off from the Android ecosystem, and an early server chip from Applied Micro that’s not publicly available. Both use custom CPU cores. Juno, by contrast, is built from ARM’s own licensable components, specifically those from the company’s suite of mobile IP.  Any firm with the wherewithal to pay the price of entry can now buy a Juno development board from ARM and test with working hardware.

The Juno development board. Source: ARM.

Both the Juno motherboard, pictured above, and the SoC that powers it were created by ARM. They’re intended to be functionally correct and a good representation of a potential mobile system, but like many development platforms, Juno isn’t meant to deliver best-in-class performance or efficiency.

Logical block diagram of the Juno system on a chip. Source: ARM.

Still, it’s interesting to see how ARM would go about building an SoC comprised of its own IP. In some respects, I think Juno’s component mix may be more sensible than the one we’ll see in end-product chips from ARM’s partners. For instance, Juno combines two relatively large (for a phone or tablet) Cortex-A57 CPU cores with four lightweight Cortex-A53 cores. This contingent of big and little cores probably better fits consumer usage patterns than the eight-core configs currently on the market.

Juno also includes ARM’s own Mali-T624 graphics processor, the mobile-focused CCI-400 uncore, dual DDR3 memory controllers, and a Cortex-M3 “system controller” block that handles SoC-level power management. Not shown in the diagram above is a low-bandwidth interface that can connect to an external FPGA logic tile. The FPGA can be programmed to emulate external devices, allowing another degree of hardware flexibility for testing.

One of the first fruits of Juno’s existence comes from the the folks at Linaro, who help create much of the core open-source software for ARM-based devices. (Linaro is a not-for-profit organization supported by ARM and a range of ARM partners; Qualcomm and MediaTek are the most recent to sign on.) Linaro has had the Juno board in its hands for “a little over a month” and is today announcing the release of a port of Google’s Android Open Source Project to the 64-bit ARM ISA. This port is essentially an early-access version of Android L for 64-bit hardware. Along with Juno, it should enable SoC and device makers to begin building and customizing their solutions right away, before their own SoCs are etched into silicon.

To give one example of the sort of thing Linaro contributes to the Android ecosystem, this first 64-bit release from Linaro already supports most optimal form of ARM’s big.LITTLE power-management scheme, known as Global Task Scheduling. In theory, GTS is the most efficient means of distributing tasks to the appropriate cores in order to maximize power efficiency, but it’s also a fairly sophisticated form of asymmetric multiprocessing that requires the explicit support of the Linux task scheduler. We’re talking about modifying a low-level core Linux kernel component, so it’s a non-trivial affair. Linaro has built GTS support into its first 64-bit release, and device makers can choose to adopt it if they wish. It’s possible Linaro’s changes could be incorporated to “upstream” projects like AOSP or even the core Linux kernel at some point in the future.

Linaro does monthly software releases, so this first 64-bit implementation of AOSP will be getting more refinements along the way as the developers get more time with Juno and the Android L preview. The folks at Linaro are eyeing OpenSSL and the cryptography instructions in ARMv8 as a potential target for optimizations. Adding hardware encryption support could boost performance and improve battery life across a broad swath of the Android ecosystem.

Meanwhile, ARM itself is using Juno in the development of drivers for its Mali series of GPUs.

I can’t help but think Juno would also, ahem, be an interesting platform for, say, a hardware review site to use in testing ARM’s 64-bit CPU cores against the competition. Just, you know, throwing that out there. Who knows what might happen next?

Comments closed
    • DarkMikaru
    • 5 years ago

    Is ARM really powerful enough for enterprise anyway? Sure, it powers our phones and tablets just fine. But hows the performance with video encoding, or 3D Modeling, photo editing, encoding?? Don’t get me wrong, I’m for anything that advances our computing cause but can ARM really compete has yet to be seen.

    *To clarify, compete in the Desktop / Server / Enterprise space.. Phones & Tablets are on lock.

      • windwalker
      • 5 years ago

      Most of enterprise computing is not heavy lifting.

      • chuckula
      • 5 years ago

      [quote<]Is ARM really powerful enough for enterprise anyway? [/quote<] A certain class of server is just there to barf up static web pages that are held in a memory cache. For that type of workload, these chips are certainly OK, although not necessarily "better" or "worse" than existing Intel or AMD low-end server parts. For everything else, ARM would need to come out with high performance designs that don't exist right now. In the process, you would see ARM's much vaunted low-power consumption numbers get tossed right out the window as ARM tried to catch up with x86.

        • NeelyCam
        • 5 years ago

        [quote<]ARM would need to come out with high performance designs that don't exist right now.[/quote<] I wonder if Intel's/AMD's high performance architecture patent portfolios will cause an insurmountable problem for ARM to ever get there Not to mention Intel is waging an intense war to try to keep ARM from entering the market, from architecture R&D, manufacturing R&D, software support... Kind of a huge mountain to climb, especially when [url=http://blogs.wsj.com/digits/2014/06/24/samsung-nvidia-shy-away-from-server-chip-battle/<]a partner after a partner is throwing in the towel[/url<]

          • raddude9
          • 5 years ago

          [quote<]I wonder if Intel's/AMD's high performance architecture patent portfolios will cause an insurmountable problem for ARM to ever get there[/quote<] Classic NPFUD! It won't. But even if it did impede it a little, it wouldn't stop AMD or Intel, both of which licence and make ARM chips by the way, from using their own IP to improve their ARM cores. [quote<]Not to mention Intel is waging an intense war to try to keep ARM from entering the market, from architecture R&D, manufacturing R&D, software support[/quote<] The battleground right now is the tablet market where Intel is paying companies to use it's chips in tablets. It's their way of keeping more profitable ARM chips out of the hands of consumers, thus reducing the money ARM has to innovate in the future. Classic Intel. [quote<]especially when a partner after a partner is throwing in the towel[/quote<] Chillax. Samsung never said they were developing server chips, and Nvidia is focusing on using the HPC route to sell more expensive chips.

            • NeelyCam
            • 5 years ago

            [quote<] it wouldn't stop AMD or Intel, both of which licence and make ARM chips by the way, from using their own IP to improve their ARM cores.[/quote<] I know AMD took this sad (yet cheap) path to new architectures, but why would Intel do something silly like that when they have a superior architecture in x86? It's dumb to pay ARM royalties when you can make a better chip using x86 royalty-free. [quote<]The battleground right now is the tablet market where Intel is paying companies to use it's chips in tablets.[/quote<] Intel charging for the tablet chip but monetarily helping with platform redesign and peripheral components is kinda similar to AMD charging $X for a GPU, then giving a bunch of games "for free", trying to steal sales from NVidia. [quote<]Samsung never said they were developing server chips, and Nvidia is focusing on using the HPC route to sell more expensive chips.[/quote<] Doesn't change the fact that the are no even remotely credible ARM server players in the game now, except maybe AMD, and we all know how well AMD is doing against Intel in servers...

            • raddude9
            • 5 years ago

            [quote<]I know AMD took this sad (yet cheap) path[/quote<] It's only sad if you're an intel fan. [quote<]a superior architecture in x86[/quote<] x86 is an instruction set, not an architecture. And if you talking instruction sets the 64-bit ARM instruction set is the most well designed and well thought out 64bit instruction set you can get. [quote<]is kinda similar to AMD charging $X for a GPU, then giving a bunch of games "for free"[/quote<] Not really, you and I both know that Intel is buying it's way to tablet marketshare using profits from it's desktop and server divisions in an effort to keep ARM out of the high-margin segments of the CPU market. [quote<]Doesn't change the fact that the are no even remotely credible ARM server players in the game now[/quote<] You think Samsung or Nvidia were credible server chip makers? What "players" would ARM need for you to deem them credible exactly?

            • just brew it!
            • 5 years ago

            IIRC Intel got out of the ARM business a few years back. Sold off their ARM division to Marvell.

          • just brew it!
          • 5 years ago

          AMD is jumping on the ARM bandwagon. So AMD’s ARM offerings (at least) will be able to utilize some of that IP.

        • albundy
        • 5 years ago

        would be nice to see a 3rd x86 competitor like back in the day when cyrix was around.

          • Stickmansam
          • 5 years ago

          VIA is still around you know

            • HisDivineOrder
            • 5 years ago

            Not that anyone notices.

      • the
      • 5 years ago

      That’s more workstation usage scenario with video encoding and 3D modeling. The short answer is ‘no’ for workstation.

      Server usage can be rather niche. Not much processing power is needed for a front end web server and the hype around ARM is predicting a take over of that niche. Stepping back and looking at the first wave of devices, ARM could dominate in this large but niche use in the data center. Following the hype to take over the entire data center isn’t going happen anytime soon. Software and RAS needs to evolve on the platform alongside performance.

      • just brew it!
      • 5 years ago

      All the compute-intensive tasks you’ve mentioned scale well to multiple cores. Ditto many server workloads. They don’t need to match x86’s single-thread performance to be competitive in these areas.

      • Deanjo
      • 5 years ago

      [quote<]But hows the performance with video encoding, or 3D Modeling, photo editing, encoding?? [/quote<] In many cases they are on par with x86. In some cases they can be faster. Nearly everything you list there is dependent on graphics performance now days. Most ARM SoC have dedicated DSP's for items like video encoding. Photo editing is transitioning to doing much of their work via the GPU and 3D modeling again is primarily GPU dependent.

    • LoneWolf15
    • 5 years ago

    This looks like it has the possibilities of a beautiful router/firewall development platform, not just regular computing/tablet/mobile devices (sadly, dev isn’t my thing).

    Scott, can we have backplane pictures of the board?

      • Damage
      • 5 years ago

      I don’t have a board on hand to photograph, sorry.

    • ronch
    • 5 years ago

    1. I’m surprised it took them this long to come out with this.

    2. This is great, as this could signal the end of the Wintel platform. We’ve had a lot of fun with Wintel but Intel’s monopoly is just getting a little tiresome. If there’s something I don’t like about this, it’s that Google is the one having a shot at dominance… and it’s not like they haven’t got it in spades already.

    3. But then, every big tech company is spying on us anyway. And we can’t expect a guppy to dominate the pond. It takes a big fish.

    4. Anyone who still thinks ARM is not interested in entering the desktop space needs to go out and have some fresh air.

    5. [quote<]Both the Juno motherboard, pictured above, and the SoC that powers it were created by ARM. They're intended to be functionally correct and a good representation of a potential mobile SoC [/quote<] The Juno motherboard and the SoC ... a good representation of a potential mobile SoC? Huh? And, um, of course it HAS to be functionally correct, doesn't it? 6. Second paragraph: [quote<]complete with an eight-core custom SoC[/quote<] Then further down... [quote<]Juno combines two relatively large (for a phone or tablet) Cortex-A57 CPU cores with four lightweight Cortex-A53 cores[/quote<] 2 + 4 = 8?

      • Damage
      • 5 years ago

      Fixed those typos. Thanks.

      • maxxcool
      • 5 years ago

      Mom dad and jr.. sure. but enterprise will not embrace ARM desktops for at least another decade.

      Arm will slowly eat at the edges until MS gets of its ass and does windows on ARM correctly. Then enterprise will adopt ARM more readily…With Ms’s help. Sad is it is to say, For day to day uses, data entry, records processing that is not automated, office work and non-engineering work MS will maintain its grip for a while to come.

      While google, redhat and suse like to tout that some obscure town in germany dumps all windows apps for linux… desktop enterprise linux still is not a viable thing for ‘most’ fortune 500’s. And even when they do..then they *still* run some apps in WINE because porting a exising win-app or adopting a linux app replacement is too costly.

        • windwalker
        • 5 years ago

        Not desktops, but thin clients.
        Most enterprise software has always had poor usability so making it all web based should be workable.

          • maxxcool
          • 5 years ago

          Yeah, that is a pretty good use for arm, thin clients would be pretty decent at any web\client to server services usage.

        • Ringofett
        • 5 years ago

        In my corner of the structural components industry, Windows has a total grip on engineering as well. It’s the preferred platform for the industry-specific software, plus there’s been a couple upstart firms that have come up with innovative hardware that boost production productivity and safety — but not being software people, they develop for the OS they know, Windows, versus trying to develop a new product AND learn a whole new OS.

        But the aforementioned engineering software suite means even our server is Windows.

        Now, corporate’s IT guys I believe use linux for other servers and in networking roles, but that’s strictly it here.

        I always figured I’d see linux take off more in new, modern POS equipment, but that hasn’t panned out; user interface wins again, because I see iPad’s being used in those sorts of roles.

        • just brew it!
        • 5 years ago

        Many business apps are already browser/web based. Even the slow, outdated, single-core ARM CPU on the Raspberry Pi is capable of running a full-blown office suite (OpenOffice), albeit rather sluggishly. For routine office use, I could see ARM desktops (or at least thin clients) grabbing a small but non-trivial slice of the market in the next few years.

        • demani
        • 5 years ago

        The thing is “Why?”. Intel is getting close performance wise, the market has shown that people are fine with the tradeoffs on the hardware now and prefer the x86 version of Windows. And Microsoft has a particularly poor track record of supporting any non-x86 ISA (Alpha and PPC come to mind). I’m not saying it can’t happen, but I don’t think it would happen until Intel’s process advantages stop giving it tools to beat performance down into the power sipping ranges that ARM is known for. And because the also operate at the higher end (the Xeon lineups) they can attack a broader segment of the market anyway.

      • NeelyCam
      • 5 years ago

      [quote<]4. Anyone who still thinks ARM is not interested in entering the desktop space needs to go out and have some fresh air.[/quote<] Of course they are interested, because they like money. Too bad they'll have a helluva difficult time to try to break into that market until full Windows supports ARM. Linux is for neckbeards, and won't be a mainstream OS until maybe 2040.

        • Farting Bob
        • 5 years ago

        [quote<]Linux is for neckbeards, and won't be a mainstream OS until maybe 2040.[/quote<] But this is the year of Linux! Also, 2013 was also the year of Linux! And 2012! And every year since the early 90's...

          • just brew it!
          • 5 years ago

          I don’t expect it to ever have a commanding share (unless you count Android). But at this point, I do expect it to hang in there, and expand into new market niches.

        • l33t-g4m3r
        • 5 years ago

        It’s pretty mainstream right now. First we have android, which is based off linux. Second, linux mint is making a lot of headway from the XP to W8 transition. A buddy of mine just installed mint, and he says he’s never going back. That’s how good it is. Close minded windows fanboys have no clue how big linux really is, and it’s going to instantaneously overtake them. They’ll wake up one day, and not know what world they’re living in, because everything will be running linux or some form of it. MS has stagnated too long to be relevant anymore.

          • NeelyCam
          • 5 years ago

          [quote<]It's pretty mainstream right now. First we have android, which is based off linux. [/quote<] I guess I should've been more clear - I meant Linux on Desktop. [quote<]Second, linux mint is making a lot of headway from the XP to W8 transition.[/quote<] Cool! Has it already reached the same sort of massive market share Windows Phone has..? [quote<]A buddy of mine just installed mint, and he says he's never going back.[/quote<] I installed KDE a while ago, saw that it's sh*t, wiped the HDD and installed XP instead. [quote<]MS has stagnated too long to be relevant anymore.[/quote<] So what was the Linux on Desktop market share again..?

            • just brew it!
            • 5 years ago

            You probably installed KDE shortly after the KDE 3 to KDE 4 transition. It did indeed suck quite badly then. FWIW these days it is my preferred desktop environment.

          • ronch
          • 5 years ago

          Right. Many people say Linux ‘just works’… Well, not in my experience. It’s more like, ‘It either works or it doesn’t. and even if it does you’d have to live with visible bugs or reduced hardware functionality. Once tried Linux Mint KDE and actually counted around 30 bugs (no, seriously). Windows has bugs too, but being a Windows user since Windows 95 I can honestly say these so-called bugs were either pretty discrete or affect only a few users or niche cases. Even now, I check our Linux distros once in a while (particularly Debian forks since I don’t really want or care much for distros like SUSE or Fedora), but there always seems to be something that prevents me from using it long term. For example, SSD support is lacking and I’ve read that TRIM support is still in its infancy. Not exactly confidence-inspiring.

          I know a lot of people here swear by Linux, and it’s a workable OS especially if you’re at home with the console, but being free doesn’t justify the bugs or having to open the hood, edit the code and fix whatever’s broken in there. If it works for you, I’m happy for you. But it hasn’t for me (as it hasn’t with many other folks I know).

          • Arclight
          • 5 years ago

          When Steam OS was announced i actually tried to use some versions of GNU/Linux and was dissapointed every time. It’s leagues behind windows in terms of user GUI or driver compatibility. It was a hassle to do the slightest thing and it’s hard to find tutorials. It just has a steeper learning curve thanWindows and then there is the issue of software which is way behind Windows (talking about support for it).

        • just brew it!
        • 5 years ago

        With their recent shifts in desktop UI paradigm Microsoft has rendered moot one of the barriers (lack of familiarity with the UI) to desktop Linux adoption. Valve is working on another one (lack of current games).

        Desktop paradigm aside, for many use cases, all that will matter going forward is whether you’ve got a standards-compliant web browser; Firefox and Chrome are both available for Linux.

Pin It on Pinterest

Share This