Stories
Slash Boxes
Comments

SoylentNews is people

posted by janrinok on Saturday March 15 2014, @11:47PM   Printer-friendly
from the more-lock-in-is-just-what-we-needed dept.

FuckBeta writes:

"Guido Stepko reports - In an GOLEM interview at CEBIT 2014 fair, Frank Kuypers, technical account manager at INTEL corp., proudly presented a new feature in INTEL processors, called "hooks", beginning with the new 2014 "Merrifield" 64 bit SoC chip generation.

In the Intel network only mobiles with certain Android versions are allowed to use certain functionalities. If you then replace your Android version, e.g. by a free Cyanogenmod Android kernel, not only some chips would stop working, e.g. LTE/UMTS, but also mails from your employer would be blinded out, because now the processor itself would 'classify' the new software as 'risk'.

Now, beginning with the new 2014 power efficient mobile "Merrifield" processor generation, this functionality will be used to lock the processor for certain OS'es or OS versions. Whether there will be a SDK or use of this 'functionality' will be kept a secret, still is undecided, Kuypers said.

Ryan O'Dell sees a potential abuse of the technology: "You'll buy a computer from a shop with Windows OS and not be able to change to Linux or another OS in the future. You may be able to buy the processor unlocked for a sum. With mobile phones/tablets it can be worse with phone networks also potentially have a lock-in. It's a disaster for the consumer"

Google translation from German: (Google)

 
This discussion has been archived. No new comments can be posted.
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
  • (Score: 2, Interesting) by jayjay.br on Sunday March 16 2014, @12:40AM

    by jayjay.br (1849) on Sunday March 16 2014, @12:40AM (#17023)

    Right, we used to do this before we had IntelVT and AMD-V. It is called binary translation. The 'ring 0' operations in the guest (where the 'feature' would be implemented) actually runs in ring 1 as 'virtual 0'. Reverting to this actually would be a step backwards, because hardware-assisted virtualization dramatically improved performance. But if it is needed it will surely happen.

    Also, yeah, it is supposed to be for mobile, and maybe for desktops, but never for servers!

    Starting Score:    1  point
    Moderation   +1  
       Interesting=1, Total=1
    Extra 'Interesting' Modifier   0  

    Total Score:   2  
  • (Score: 2) by ls671 on Sunday March 16 2014, @02:03AM

    by ls671 (891) Subscriber Badge on Sunday March 16 2014, @02:03AM (#17036) Homepage

    I have the vmx flag on intel and I use "hardware-assisted virtualization" with qemu.

    There still seems to be an abstraction layer there, probably handy for portability. I think I remember playing with qemu command line parameters, you can expose more native cpu flags to the guest. I think you can even expose fake flags to the guests and that qemu will emulate the flag behavior.

    This is with the default qemu settings.
    Host:

    processor : 0
    vendor_id : GenuineIntel
    cpu family : 6
    model : 15
    model name : Intel(R) Core(TM)2 Quad CPU Q6600 @ 2.40GHz
    stepping : 11
    microcode : 0xb6
    cpu MHz : 2399.804
    cache size : 4096 KB
    physical id : 0
    siblings : 4
    core id : 0
    cpu cores : 4
    apicid : 0
    initial apicid : 0
    fpu : yes
    fpu_exception : yes
    cpuid level : 10
    wp : yes
    flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx lm constant_tsc arch_perfmon pebs bts rep_good nopl aperfmperf pni dtes64 monitor ds_cpl vmx est tm2 ssse3 cx16 xtpr pdcm lahf_lm dtherm tpr_shadow vnmi flexpriority
    bogomips : 4799.60
    clflush size : 64
    cache_alignment : 64
    address sizes : 36 bits physical, 48 bits virtual

    Guest:
    processor : 0
    vendor_id : GenuineIntel
    cpu family : 6
    model : 2
    model name : QEMU Virtual CPU version 1.4.1
    stepping : 3
    cpu MHz : 2400.120
    cache size : 4096 KB
    fdiv_bug : no
    hlt_bug : no
    f00f_bug : no
    coma_bug : no
    fpu : yes
    fpu_exception : yes
    cpuid level : 4
    wp : yes
    flags : fpu de pse tsc msr pae mce cx8 apic mtrr pge mca cmov pse36 clflush mmx fxsr sse sse2 syscall nx lm up pni cx16 hypervisor lahf_lm
    bogomips : 4800.24
    clflush size : 64
    power management:

    --
    Everything I write is lies, including this sentence.
    • (Score: 1) by jayjay.br on Sunday March 16 2014, @02:44AM

      by jayjay.br (1849) on Sunday March 16 2014, @02:44AM (#17044)

      You're absolutely correct about qemu. There are still remains of the emulation layer originally meant to be. Ring 3 stuff runs directly and ring 0 calls for the "hardware assisted emulation". For VMware though, is quite a different process, and it exposes almost every bit of the processor, except for the root mode privileged execution. This also makes it somewhat of a pain for migrating VMs among different families of processors.

      Nice read about this: http://www.vmware.com/files/pdf/VMware_paravirtual ization.pdf [vmware.com]