Archive

Posts Tagged ‘armv8’

Linaro 15.04 Release with Linux 4.0 and Android 5.1

May 1st, 2015 1 comment

Linaro 15.04 has been released with Linux 4.0 (baseline), Linux 3.10.74 and 3.14.39 (LSK), and Android 5.1.1.

Other noticeable changes include support for the new DragonBoard 410c 96boards compliant board, the addition of A80 Optimusboard (Allwinner A80) to Android Kitkat build, Hisilicon D01 support added to the Debian installer, and support for Ubuntu ARM64 Gnome rootfs.

Highlights of the release:

  • Linux Linaro 4.0-2015.04
    • updated linaro-android topic: aosp/android-3.18 branch has been merged
    • GATOR topic: version 5.20.1
    • updated integration-linaro-vexpress64 topic by ARM LT (FVP Base and Foundation models, and Juno support)
    • updated topic from Qualcomm LT (IFC6410 and DB410c boards support):
      • Resource Power Manager (RPM) – MSM Shared Memory Driver (SMD) driver
      • quite some changes under drivers/gpu/drm/ related to adv7511 and adv7533 support
      • ASoC support for QCOM platforms
      • external Connector Class (extcon) support (used for USB VBUS and ID detection)
  • Linaro builds of AOSP 15.04
    • baseline updated to android-5.1.1_r1
    • updated Nexus 10 CI for LAVA testing
    • added Optimus A80 Android Kitkat build
  • Linaro OpenEmbedded 2015.04
    • removed stress recipe in favor of oe-core recipe
    • updated linux-linaro to ll_20150422.0 (based on 4.0)
    • libevent-fb: OE-core updated to 2.0.22, fix require statement
    • APM mustang boot failure was tracked down to using ‘arm64′ as U-Boot arch for the initramfs header while the vendor U-Boot 2013.04 expects ‘arm’.
    • upstreaming – strace: fix build for aarch64; libgpg-error 1.18: simplify tuple handling and add armv8b support
  • Linaro Ubuntu 15.04 – updated packages: LSK 3.10.74/3.14.39 and linux-linaro 4.0 kernels
  • U-Boot: upstream fastboot support
  • Add HiSilicon D01 platform support to Debian installer
  • Updated android-build job to work with docker slaves
  • Cleaned up ILP32 build job
  • LSK: enable debug options on regular builds
  • linux-linaro: ll-fold.sh script fixed to work correctly with newer git versions
  • CI bring up: tshark board Android member build
  • Added Ubuntu arm64 gnome rootfs
  • Added D01 platform to Coresight CI loop

Check out https://wiki.linaro.org/Cycles/1504/Release for a list of known issues, and further release details about the LEB, LMB (Linaro Member Builds), and community builds, as well as Android, Kernel, Graphics, Multimedia, Landing Team, Platform, Power management and Toolchain components.

Digg This
Reddit This
Stumble Now!
Buzz This
Vote on DZone
Share on Facebook
Bookmark this on Delicious
Kick It on DotNetKicks.com
Share on LinkedIn
Bookmark this on Technorati
Post on Twitter

Linux 4.0 Release – Main Changes, ARM and MIPS Architectures

April 15th, 2015 6 comments

Linus Torvalds “Ima Sheep” released Linux Kernel 4.0 on Sunday:

So I decided to release 4.0 as per the normal schedule, because there really weren’t any known issues, and while I’ll be traveling during the end of the upcoming week due to a college visit, I’m hoping that won’t affect the merge window very much. We’ll see.

Linux 4.0 was a pretty small release both in linux-next and in final size, although obviously “small” is all relative. It’s still over 10k non-merge commits. But we’ve definitely had bigger releases (and judging by linux-next v4.1 is going to be one of the bigger ones).

Which is all good. It definitely matches the “v4.0 is supposed to be a_stable_ release”, and very much not about new experimental features etc. I’m personally so much happier with time-based releases than the bad old days when we had feature-based releases.

That said, there’s a few interesting numerological things going on with 4.0. Looking at just the statistics in git, this release is not just when we cross half a million commits total, but also cross the 4 million git object limit. Interestingly (if you look for numeric patterns), Linux 3.0 was when we crossed a quarter million commits and 2 million git objects, so there’s a nice (and completely unintentional) pattern there when it comes to the kernel git repository.

[ Another quick historical numerological footnote: the old historical BK tree was getting close to the 16-bit commilt limit that BK originally used to have. So that whole “quarter of a million commits” is actually quite a lot. During all of the BK years we only got 65k commits. Of course, we only used BK for three years, and we’ve now been on git for almost exactly ten years, but still – it shows how the whole development process has really sped up a _lot_ ]

Feature-wise, 4.0 doesn’t have all that much special. Much have been made of the new kernel patching infrastructure, but realistically, that not only wasn’t the reason for the version number change, we’ve had much bigger changes in other versions. So this is very much a “solid code progress” release.

Go get it and enjoy,

Linus “we’re all sheep” Torvalds

Linux 3.19 brought improvement to btrfs (raid), the network stack, added ARM Coresight, device tree overlays support, and more.

Some key changes made to Linux 4.0 include:

  • pNFS (Parallel NFS), UBIFS, F2FS and BTRFS File Systems improvements
  • Live Kernel Patching – Install kernel updates without rebooting
  • Intel Quark x86 SoC support
  • Various patches to improve Linux running on a  Playstation 3
  • Open source AMD Radeon driver supports DisplayPort Audio and improves fan support

Some of the new features and improvements specific to the ARM architecture include:

  • Allwinner:
    • A20 – PS/2 Controller
    • A31 – IR receiver
    • A31s – Bring-up sharing majority of drivers with A31, pinctrl driver
    • A80 – MMC
    • All SoCs – LRADC Input driver, CPUFreq, PWM Driver
    • AXP209 power button input driver
    • New boards and devices:  CSQ CS908, LeMaker Banana Pro, Chuwi V7 CW0825, Rikomagic mk802, Rikomagic mk802ii, Rikomagic mk802_a10s, MarsBoard A10, Hyundai A7HD
  • Rockchip
    • Fixes for rk808 regulator
    • Watchdog fix
    • Add Rockchip timer for RK3288
    • HDMI output enabled on rk3288-firefly and rk3288-evb
    • Disable GMAC by default
  • Amlogic – pinctrl driver for Amlogic Meson SoCs
  • Mediatek
    • Regulator driver for Mediatek MT6397
    • Added watchdog driver
    • Added Mediatek MT8173 64-bit processor
  • ARM64
    • New processors: Exynos 7, Freescale LS2085A, and Tegra 132 (Denver)
    • Various fixes for ARM64 including UEFI and KVM code.
  • Preparation work for Atmel AT91 support for multiplatform
  • Other new platforms – Alphascale ASM9260, Marvell Armada 388, CSR Atlas7, TI Davinci DM816x, Hisilicon HiP01, ST STiH418, and Conexant Digicolor (CX92755).

There has also been some interesting changes for the MIPS architecture:

  • Fixes for KVM support
  • Support for MIPS R6 processors
  • Preliminary support for Cavium Octeon 3 SoCs which feature up to 48 MIPS64 R3 cores with FPU and hardware virtualization

A more detailed changelog for Linux 4.0 will soon be available on Kernelnewbies.org, and once it’s up you may also want to have a look at their ARM architecture and drivers sections for more details about changes related to ARM, MIPS and other platforms. I’ve also generated a complete Linux 3.19 vs Linux 4.0 changelog (3.4MB) with git (comments only, no code).

Digg This
Reddit This
Stumble Now!
Buzz This
Vote on DZone
Share on Facebook
Bookmark this on Delicious
Kick It on DotNetKicks.com
Share on LinkedIn
Bookmark this on Technorati
Post on Twitter

HPC Performance & Power Usage Comparison – Intel Xeon E3 vs Intel Atom C2720 vs Applied Micro X-Gene 1 vs IBM Power 8

April 14th, 2015 6 comments

Last year, the CERN published a paper comparing Applied Micro X-Gene (64-bit ARM) vs Intel Xeon (64-bit x86) Performance and Power Usage, and they’ve now added IBM Power 8 and Intel Atom Avoton C2750 processor to the mix in a new presentation entitled “A look beyond x86: OpenPOWER & AArch64“.
ARM_x86_Power_8_Test_Systems
So four systems based on Intel Xeon E3-1285L, Intel Atom C2750, Applied Micro X-Gene 1, and IBM Power 8 were compared, all running Fedora 21, except the HP Moonshot 1500 ARM plarform running Ubuntu 14.04 and an older kernel. All four systems use gcc 4.9.2, and Racktivity intelligent PDUs were used for power measurement.

I’ll just share some of their results, you can read the presentation, or go through the benchmark results to find out more.

HEP-SPEC06_Results

HEP-SPEC06 Benchmark (Click to Enlarge)

HEP-SPEC06 is a new High Energy Physics (HEP) benchmark for measuring CPU performance developed by the HEPiX Benchmarking Working Group, and here it’s not surprising to see the low power solutions under-perform the more powerful Intel Xeon and Power 8 processors, with the latter taking the crown.

Geant_4_ParFullCMS

Geant 4 ParFullCMS (Click to Enlarge)

Geant 4 simulates the passage of particles through matter, something that you would expect the CERN to do regularly. Intel Xeon E3 outperforms  IBM Power8 processor here.

But let’s move on to power consumption, and performance per watt.

Idle Power Consumption (Click to Enlarge)

Idle Power Consumption (Click to Enlarge)

IBM OpenPower 8 has a much higher power consumption than other systems, and HP Moonshot ARM 64-bit X-Gene 1 consumes more than both Intel servers. The chart under full load (not shown here) also shows a similar pattern.

HEP_SPEC06_Per_Watt

HEP-SPEC06 per Watt (Click to Enlarge)

When it comes to performance per watt however, both HP Moonshot ARM and Power 8 systems are the least efficient here, and Intel systems provide the best ratio. Bear in mind that X-Gene 1 is manufactured with a 40nm process, while Applied Micro X-Gene 2  and 3 will be manufactured using 28nm and 16 nm FinFET processes, so some large efficiency gains could be expected here.

We may find out soon, as the CERN expects to add these two new processors, as well a Cavium ThunderX to their benchmarks in the future.

Thanks to David for the tip.

Digg This
Reddit This
Stumble Now!
Buzz This
Vote on DZone
Share on Facebook
Bookmark this on Delicious
Kick It on DotNetKicks.com
Share on LinkedIn
Bookmark this on Technorati
Post on Twitter

Relative Performance of ARM Cortex-A 32-bit and 64-bit Cores

April 9th, 2015 12 comments

Many people assume newer processors will be faster, or that 64-bit processor will provide a performance boost compared to 32-bit processors, but the reality can be quite different, and I’ve decided to have a look at ARM Cortex-A cores using ARMv7 (32-bit) and ARMv8 (64-bit) architecture, and see what kind of integer performance you can expect from each at a given frequency. To do so, I’ve simply use DMIPS/Mhz (Dhrystone MIPS/Megahertz) values listed on Wikipedia.

Vertical Scale: DMIPS / MHz

Vertical Scale: DMIPS / MHz


Drystone benchmark has no floating-point operating, so it’s a pure integer benchmark. I’m only looking at ARM core here, and once integrated in an SoC, other parameters like memory bandwidth, amount of cache,  GPU, etc.. will greatly affect the overall system performance. The figure above are per MHz, and it does not mean for example that a Cortex A5 processor will be slower than a Cortex A7 processor, as can be seen by the comparison between Amlogic S805 (4x Cortex A5) and Broadcom BCM2835 (4x Cortex A7), which shows the Amlogic processor is about 40% faster due to higher clock speed.

With that in mind, it can be seen than you may not expect all recent Cortex A53 processors to outperform existing Cortex A15 and A17 processors, and in some case even Cortex A9 processors, and the real performance benefit with 64-bit cores only start to show with Cortex A57, and especially Cortex A72 cores which is some cases could be twice as fast as Cortex A15 cores. The red zone on top of some bars represents the possible performance variation due to different implementations of the cores.

ARMv8 also brings some other improvement such as additional cryptographic extensions, an increase in the number of SIMD/floating point, and general purpose registers, and more, as shortly explained in that article. All of these should also deliver benefits provided the firmware and applications support them.

Digg This
Reddit This
Stumble Now!
Buzz This
Vote on DZone
Share on Facebook
Bookmark this on Delicious
Kick It on DotNetKicks.com
Share on LinkedIn
Bookmark this on Technorati
Post on Twitter

Rockchip RK3368 64-bit ARM Android Boxes Are Coming Soon

April 8th, 2015 29 comments

Yesterday, Rockchip had an event for their Atom x3-C3230RK SoCs (previously known as Sofia)  developed in collaboration with Intel, and upcoming Android 5.1 tablets and smartphones based on the new platform, as well as progress with their first 64-bit ARM SoC (RK3368) that is now integrated into Android TV boxes, albeit still in development. Padnews wrote a summary of the event, but in this post I’ll focus on Rockchip RK3368 hardware.

Rockhcip RK3368 TV Box Board (Clikc to Enlarge)

Rockchip RK3368 TV Box Board (Click to Enlarge)

One of the first product will be X6 TV box. Padnews did not have a readable picture with the specs, but Eric of Geekbuying also went to the event, and kindly provided a clear picture with the specs (Mali-600MP is however not the GPU used in RK3368):

  • SoC – Rockchip RK3368 octa core Cortex A53 processor with PowerVR G6110 GPU with support for OpenGL ES 1.x/2.0/3.x, OpenGL 3.2, DirectX 9.3, OpenCL 1.2 EP, and Renderscript
  • System Memory – 1 or 2 GB RAM
  • Storage – 8 to 16 GB eMMC flash
  • Video Output – HDMI, composite RCA
  • Audio Output – HDMI, stereo RCA
  • Connectivity – Ethernet, Wi-Fi 802.11 b/g/n, Bluetooth 4.0
  • USB – 1x USB host port, 1x USB OTG port (full size)
  • Dimensions – 188 x 115 x 34 mm

The device will run Android 5.0.

X6_TV_BoxTraditionally, Rockchip has gone with Mali GPUs, except with RK3168, and now, RK3368 that features PowerVR G6110, the most powerful GPU of imagination’s entry-level PowerVR Series6XE family.

Don’t expect amazing performance, as RK3368 will not be quite as fast as RK3288, but it will be more cost effective. I’ve also been told X6 has been designed or the Chinese market only, and won’t be sold overseas, but that other RK3368 products are being developed for the rest of the world…

Digg This
Reddit This
Stumble Now!
Buzz This
Vote on DZone
Share on Facebook
Bookmark this on Delicious
Kick It on DotNetKicks.com
Share on LinkedIn
Bookmark this on Technorati
Post on Twitter

Tegra X1 Chromebooks Likely in the Works

March 30th, 2015 No comments

When Nvidia announced Tegra X1 processor, the company’s main focus appeared to be the automotive market as it introduced Drive PX and CX board for this very market. Since then, Tegra X1 based SHIELD console was unveiled, and I’ve been informed some development activity related to Tegra X1 (Model: Tegra210 / T210) was taking place in Chromium and especially Coreboot source code.

Tegra_X1_Chromebook_Code There appears to be two hardware platforms based on Tegra X1: smaug and foster which could end-up being Chromebooks or Chromeboxes likely with 4GB RAM.

Thanks to David for the information.

Digg This
Reddit This
Stumble Now!
Buzz This
Vote on DZone
Share on Facebook
Bookmark this on Delicious
Kick It on DotNetKicks.com
Share on LinkedIn
Bookmark this on Technorati
Post on Twitter

Howchip is Teasing ExSOM-7420SB Development Board Based on Samsung Exynos 7420 Processor

March 27th, 2015 2 comments

We’ve already seen a few Cortex A53 boards announced in the last few months with Nobel64, as well as Hikey & DragonBoard 410c 96Boards, but none of them are based on the more powerful Cortex A57 cores. Howchip is going to change that with the upcoming ExSOM-7420SB single board computer featuring Samsung Exynos 7420 processor used in Galaxy S6 smartphone.

ExSOM7420 Block Diagram

ExSOM7420 Block Diagram

The company has released very few details about the board, except the block diagram above that shows Exynos 7420 with 3GB LPDDR4 PoP memory, and various interfaces such as USB 2.0/3.0, UFS/eMMC, Ethernet, HDMI, MIPI DSI, Camera, as well as Wi-Fi and Bluetooth connectivity. The board will support Android 5.0 Lollipop 64-bit. No word about Linux.

That’s the video teaser, but you won’t learn much…

More details should eventually be published on ExSOM-7420SB product page. Alternatively, Hardkernel will also probably launch an ODROID-XU4? board based on the latest Exynos 7 processor in due time.

Digg This
Reddit This
Stumble Now!
Buzz This
Vote on DZone
Share on Facebook
Bookmark this on Delicious
Kick It on DotNetKicks.com
Share on LinkedIn
Bookmark this on Technorati
Post on Twitter

Gigabyte MP30-AR0 is an ARM Server Motherboard Powered by Applied Micro X-Gene 1 SoC

March 27th, 2015 14 comments

So far, it’s been pretty hard to buy ARM server motherboards for individuals, as most, if not all, products were reserved to corporate entities, but with Gigabyte MP30-AR0 server motherboard featuring the first generation Applied Micro X-Gene 64-bit ARM processor this might be about to change. [Update: As mentioned in comments I was probably wrong here, since the motherboard is listed on the Gigabyte’s B2B website, and not its B2C website].

MP30-AR0MP30-AR0 specifications:

  • Processor – AppliedMicro X-Gene 1 processor with 8 ARMv8 cores up to 2.4GHz (TDP 45W)
  • System Memory – 8 x DIMM slots, Single, dual rank UDIMM modules @ 1333/1600 NHz supported (up to 16GB)
  • Storage – 4x SATA III 6Gb/s ports + 1x SD card slot
  • Connectivity – 2x 10GbE SFP+ LAN ports (integrated), 2x GbE LAN ports (Marvell 88E1512), 1x 10/100/1000 management LAN
  • Graphics – Video Integrated in Aspeed AST2400. 2D Video Graphic Adapter with PCIe bus interface up to 1920×1200@60Hz 32bpp.
  • Expansion Slots – 2x PCIe x16 (Gen3 x8 bus) slots
  • Other Internal I/O
    • 1 x CPU fan header
    • 4x system fan headers
    • 1x USB 2.0 header
    • 2x Front panel headers
    • 1x APM strap header
    • 1x HDD back plane board header
    • 1x PMBUS header
    • 1x BMC JTAG header, 1x JTAG PLD header
    • 1x BIOS_H header
    • 1x Chassis intrusion header
    • 1x SATA DOM jumper, 1x BIOS recovery jumper, 1x ACK selection jumper
    • 1x IPMB connector
  • Rear I/Os
    • 2x USB 2.0, 1x Mini USB
    • 1x VGA
    • 1x Serial
    • 2x SFP+, 3x RJ45
    • 1x ID button with LED, 1x Power button with LED, 1x Status LED
  • Power – 1x 24-pin ATX main power connector; 2x 4-pin ATX 12V power connectors
  • Dimensions –  244 × 244 mm (microATX form factor)
MP30-AR0 Motherboard (Click to Enlarge)

MP30-AR0 Motherboard (Click to Enlarge)

The motherboard supports Ubuntu 14.04, and can also be configured with Avocent MergePoint IPMI 2.0 web interface.

Pricing information is still to be announced according to the motherboard page, and the company also integrated it into R120-P30 single socket 1U rackmount server with a 350W PSU and support for 4 hard drives.

Via Tom Cubie

Digg This
Reddit This
Stumble Now!
Buzz This
Vote on DZone
Share on Facebook
Bookmark this on Delicious
Kick It on DotNetKicks.com
Share on LinkedIn
Bookmark this on Technorati
Post on Twitter