Intel updates mysterious ‘software-defined silicon’ code in the Linux kernel
(2021/12/08)
- Reference: 1638946749
- News link: https://www.theregister.co.uk/2021/12/08/intel_software_defined_silicon_update/
- Source link:
Intel has updated the code it says allows the implementation of “software-defined silicon” (SDSi).
Chipzilla dropped some code for SDSi into the Linux Kernel in September 2021, describing it as tech that allows users to activate dormant features in silicon. The code outlined a process for enabling new features by verifying cryptographically signed licences.
A new [1]post to the Linux Kernel Mailing List from David E. Box, a Linux dev who works at Intel, explains version two of Intel’s SDSi code.
[2]
Box explained the update offers “changes to the current intel_pmt driver to give it broader support for Intel defined PCIe VSEC and DVSEC features. It moves the implementation from MFD to the auxiliary bus and creates a generic framework for enumerating the extended capabilities. It also adds support for a new VSEC, Software Defined Silicon (SDSi).”
[3]
[4]
VSEC stands for “Vendor-Specific Extended Capability,” in case you are wondering. And this [5]PCI-SIGN Engineering Change Notice (PDF) from 2015 states it “allows PCI Express component vendors to use the Extended Capability mechanism to expose vendor-specific registers.”
The Register asked Intel about SDSi in the wake of Box’s first posts on the matter. Intel declined to offer details of its plans, and even [6]told The Register SDSi might not have anything to do with an actual product and/or might never be used.
[7]Self-driving towards an IPO? Intel unveils plans for Mobileye offering
[8]Intel's recent Atom, Celeron, Pentium chips can be lulled into a debug mode, potentially revealing system secrets
[9]Intel hopes to burn newly open-sourced AI debug tech into chips
We were sceptical of that assertion because making the effort to add code to the Linux kernel without planning for it to be used is thoroughly counter-intuitive. It’s also the sort of thing that annoys key Linux maintainers: Greg Kroah-Hartman [10]fumed when researchers conducted an experiment by trying to add insecure code to the kernel and then [11]testily pointed out that 80 developers worked to clean up the resulting mess.
Intel knows better than to provoke that sort of ire. And now that it’s bothered to refine its SDSi efforts, the non-explanation for its kernel contributions is even flimsier than before.
[12]
We’ve therefore asked Intel to again explain its SDSi ambitions. If we receive a meaningful response, we’ll let you know. ®
Get our [13]Tech Resources
[1] http://lkml.iu.edu/hypermail/linux/kernel/2112.0/07851.html
[2] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_onprem/front&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=2&c=2YbCQXib9IaPyBIvI0KGaXAAAAAw&t=ct%3Dns%26unitnum%3D2%26raptor%3Dcondor%26pos%3Dtop%26test%3D0
[3] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_onprem/front&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44YbCQXib9IaPyBIvI0KGaXAAAAAw&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[4] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_onprem/front&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33YbCQXib9IaPyBIvI0KGaXAAAAAw&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0
[5] https://composter.com.ua/documents/ECN_DVSEC-2015-08-04.pdf
[6] https://www.theregister.com/2021/10/19/intel_sdsi/
[7] https://www.theregister.com/2021/12/07/mobileye_ipo/
[8] https://www.theregister.com/2021/11/16/intels_chip_flaw/
[9] https://www.theregister.com/2021/10/25/intel_controlflag/
[10] https://www.theregister.com/2021/04/21/minnesota_linux_kernel_flaws_update/
[11] https://www.theregister.com/2021/05/21/linux_5_13_patches/
[12] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_onprem/front&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44YbCQXib9IaPyBIvI0KGaXAAAAAw&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[13] https://whitepapers.theregister.com/
Chipzilla dropped some code for SDSi into the Linux Kernel in September 2021, describing it as tech that allows users to activate dormant features in silicon. The code outlined a process for enabling new features by verifying cryptographically signed licences.
A new [1]post to the Linux Kernel Mailing List from David E. Box, a Linux dev who works at Intel, explains version two of Intel’s SDSi code.
[2]
Box explained the update offers “changes to the current intel_pmt driver to give it broader support for Intel defined PCIe VSEC and DVSEC features. It moves the implementation from MFD to the auxiliary bus and creates a generic framework for enumerating the extended capabilities. It also adds support for a new VSEC, Software Defined Silicon (SDSi).”
[3]
[4]
VSEC stands for “Vendor-Specific Extended Capability,” in case you are wondering. And this [5]PCI-SIGN Engineering Change Notice (PDF) from 2015 states it “allows PCI Express component vendors to use the Extended Capability mechanism to expose vendor-specific registers.”
The Register asked Intel about SDSi in the wake of Box’s first posts on the matter. Intel declined to offer details of its plans, and even [6]told The Register SDSi might not have anything to do with an actual product and/or might never be used.
[7]Self-driving towards an IPO? Intel unveils plans for Mobileye offering
[8]Intel's recent Atom, Celeron, Pentium chips can be lulled into a debug mode, potentially revealing system secrets
[9]Intel hopes to burn newly open-sourced AI debug tech into chips
We were sceptical of that assertion because making the effort to add code to the Linux kernel without planning for it to be used is thoroughly counter-intuitive. It’s also the sort of thing that annoys key Linux maintainers: Greg Kroah-Hartman [10]fumed when researchers conducted an experiment by trying to add insecure code to the kernel and then [11]testily pointed out that 80 developers worked to clean up the resulting mess.
Intel knows better than to provoke that sort of ire. And now that it’s bothered to refine its SDSi efforts, the non-explanation for its kernel contributions is even flimsier than before.
[12]
We’ve therefore asked Intel to again explain its SDSi ambitions. If we receive a meaningful response, we’ll let you know. ®
Get our [13]Tech Resources
[1] http://lkml.iu.edu/hypermail/linux/kernel/2112.0/07851.html
[2] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_onprem/front&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=2&c=2YbCQXib9IaPyBIvI0KGaXAAAAAw&t=ct%3Dns%26unitnum%3D2%26raptor%3Dcondor%26pos%3Dtop%26test%3D0
[3] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_onprem/front&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44YbCQXib9IaPyBIvI0KGaXAAAAAw&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[4] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_onprem/front&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33YbCQXib9IaPyBIvI0KGaXAAAAAw&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0
[5] https://composter.com.ua/documents/ECN_DVSEC-2015-08-04.pdf
[6] https://www.theregister.com/2021/10/19/intel_sdsi/
[7] https://www.theregister.com/2021/12/07/mobileye_ipo/
[8] https://www.theregister.com/2021/11/16/intels_chip_flaw/
[9] https://www.theregister.com/2021/10/25/intel_controlflag/
[10] https://www.theregister.com/2021/04/21/minnesota_linux_kernel_flaws_update/
[11] https://www.theregister.com/2021/05/21/linux_5_13_patches/
[12] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_onprem/front&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44YbCQXib9IaPyBIvI0KGaXAAAAAw&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[13] https://whitepapers.theregister.com/
The plan is obvious
Flocke Kroes
Intel will sell CPUs that operate in a crippled mode. For some extra money you can buy a crypto key to uncripple one feature of one specific CPU.
Re: The plan is obvious
ecofeco
Will? They've been doing this for decades.
Re: The plan is obvious
A Non e-mouse
Or if you're a three-letter agency...
Re: The plan is obvious
Zanzibar Rastapopulous
The World Wildlife Fund?
Re: The plan is obvious
Zippy´s Sausage Factory
The more I read about Intel, the more I want to buy AMD processors...
Some things never change
Peter Gathercole
You know, this sounds suspiciously like what IBM, Amdahl and some of the larger mini-computer manufacturers were doing in the '60s, '70s and '80s with microcoded processor extensions.
Pay a fortune, have an engineer turn up with a super secret floppy disk, and get a whole load of new instructions added to your processor.
The only difference now is that we have the Internet, so they have to protect this code with cryptography, to prevent it leaking.
Re: Some things never change
imanidiot
"The only difference now is that we have the Internet, so they have to protect this code with cryptography, to prevent it leaking."
more like delay the leaking a few days in reality.
Trust free zone
That's how this comes over to me. I can think of several ways this could be abused - and I know virtually nothing about chip design.