zlacker

Pluton is not currently a threat to software freedom

submitted by foodst+(OP) on 2022-01-09 02:06:36 | 143 points 204 comments
[view article] [source] [go to bottom]

NOTE: showing posts with links only show all posts
3. transp+78[view] [source] 2022-01-09 03:14:39
>>foodst+(OP)
Background material on Pluton:

1. Xbox Security, https://www.platformsecuritysummit.com/2019/speaker/chen/

2. Azure Sphere (derived from Xbox) with Microsoft Linux kernel, OE/Yocto runtime and QEMU emulation of Pluton for CI/CD, https://www.platformsecuritysummit.com/2019/speaker/seay/

3. DMTF SPDM (PCI device firmware attestation to SoC/RoT), https://www.platformsecuritysummit.com/2019/speaker/plank/

Nov 2020 Intel announcement about Pluton, https://itpeernetwork.intel.com/intel-and-microsoft-plan-to-...

> Secure platforms anchor on a hardware Root of Trust as the foundation. Given Intel’s diverse ecosystem, our vision is to offer multiple Root of Trust options that ensure isolation of resources, keys and security assets. The partnership with Microsoft to offer Pluton will further broaden the choices available to our mutual customers.

Hopefully a future Intel SoC will include an optional FPGA-based RoT where customer hardware owners can load the open-source firmware of their choice.

Edit: Pluton will be included in upcoming Arm laptops with SoCs from the Qualcomm-Nuvia (former Apple M1) team.

◧◩◪◨⬒
30. transp+Jc[view] [source] [discussion] 2022-01-09 03:57:01
>>xyzzy_+8c
There's a direct line between the Microsoft/AMD Xbox SoC corporate contract and Pluton in 2022 Ryzen consumer CPUs, as described in the videos linked above. It's not in Intel's interest to make Microsoft-AMD designed Pluton into the exclusive silicon RoT provider for Intel CPUs.

Another candidate for "Open RoT" is Google's OpenTitan, https://opentitan.org.

Open-Source FPGA Foundation: https://osfpga.org/

◧◩◪
37. messe+od[view] [source] [discussion] 2022-01-09 04:02:21
>>heavys+Lb
> Both Windows and macOS are approaching the iOS security model that depends on manufacturers blessing what software can run on their products, and banning software they don't want users to run.

That's been said for years, and hasn't held true. I can boot a Linux kernel on my M1 macbook. Apple could easily have locked it down in exactly the same manner as their iOS/iPadOS devices, yet chose not to. I can still install whatever I want. The default state of the system has a locked down root volume. And the default behaviour is not to install untrusted software, unless you jump through a couple of hoops. Those are good defaults. Those are damn good defaults for most people. If you're running untrusted code in your webbrowser all day long, you want your base system to be as unmalleable as possible, and as untrusting as possible to third party code. But I can still work around that with almost no hassle. Homebrew still installs software as easily as it used to nearly a decade ago; it just might need the occasional --no-quarantine flag for unsigned software.

Even recently they appeared to have actively assisted in the running on non-macOS operating systems on their hardware: removing the requirement for kernel images to be in mach-O format[1].

[1]: https://twitter.com/marcan42/status/1471799568807636994

◧◩
49. zucker+Me[view] [source] [discussion] 2022-01-09 04:14:23
>>messe+sa
> The fearmongering about Pluton

Part of the reason for this "fearmongering" (if it's fair to call it that) is that Microsoft has released little information about Pluton, besides a press release. Plus, it's not like the fears are completely unfounded based on Microsoft's messaging; Microsoft's press release says Pluton is based off the Xbox[1] (and this paywalled article mentions the same thing[3]), and they've previous said the major goal of the Xbox security system is piracy prevention [2], i.e. DRM. However, I agree with the overall conclusion of the main article that it's probably not much worse that what already exists.

[1] https://blogs.windows.com/windowsexperience/2022/01/04/ces-2...

[2] https://www.platformsecuritysummit.com/2019/speaker/chen/

[3] https://ieeexplore.ieee.org/abstract/document/9354509

◧◩◪◨⬒⬓
83. judge2+Vh[view] [source] [discussion] 2022-01-09 04:44:47
>>dane-p+5f
None of these have any requirement on some TPM specification. A government can already do as much invasive monitoring as they want, either by forcing citizens to install MITM root CAs[0] or generally requiring invasive identity checks when people sign in, or just limiting what privacy-invasive devices are even allowed to be sold at all. Banks can already go "lol no web frontend for you, go use our mobile app". And neither Visa/Mastercard nor their bank partners are going to allow such strict restrictions that'll surely reduce the amount of impulsive purchases people can make, and you forget that every online payment is already hard tied to your identity via your bank / credit accounts.

> in that you are responsible for whatever is done while you are logged in, unless you report it stolen.

This is only really true for insurance purposes - for stuff like red light cams, the tickets are invalid if you weren't the one driving (which is why some newer ones snap temporary pictures of people in the driver seat in case they end up running the light).

0: https://news.ycombinator.com/item?id=20472179

◧◩◪◨⬒⬓⬔
88. transp+Si[view] [source] [discussion] 2022-01-09 04:53:36
>>mjg59+5g
not much detail, but slide 12 claims: https://www.platformsecuritysummit.com/2019/speaker/seay/PSE...

> Pluton validates and boots Security Monitor

> Security Monitor validates and boots the Linux Kernel

> Application Signatures are verified by SM and Pluton before Linux Kernel loads an application

◧◩◪◨⬒
93. gruez+ek[view] [source] [discussion] 2022-01-09 05:06:56
>>datafl+ch
>Maybe if you're talking Java versions, but not in the desktop OS space. (Look at so many old machines running Windows 7 or earlier right now, and look at long old OSes are officially supported, and how long they're still used afterward.)

see: https://news.ycombinator.com/item?id=29860320

>They take their time and meander, and it takes a while. Possibly due to corporate sluggishness, possibly due to wanting to boil the frog slowly, possibly due to wanting to test the waters for a while... who knows why. But speed isn't the main criterion.

But in this case it's not really boiling the frog because it's not really getting worse? All we know so far is that it's TPM but it's easier to update. I suppose this could be used to oppress users by patching jailbreaks faster, but the security benefits at least makes it plausible that they're not doing it as some sort of plan to oppress users.

◧◩◪◨⬒⬓⬔
98. dane-p+4o[view] [source] [discussion] 2022-01-09 05:46:31
>>judge2+Vh
Right now it is politically unthinkable for Western governments to demand people install MITM root CAs, and technically infeasible that they would re-encrypt every TLS connection (and check for encryption being layered inside the decrypted streams). (When Kazakhstan tried, they also faced resistance from software makers, but I wonder what would happen if those software makers happened to be based in the same country that was implementing this policy.)

It is much more thinkable, however, especially in 5 years, perhaps after a (false flag?) cyber-attack takes down an electricity grid in some country, that a government could prevent "insecure"/"unpatched" devices from going online. This wouldn't require any personal information to be shared with the government (at least, no more than current ISP data retention laws already require), and Microsoft would be all too happy to build support for this right into Windows for free, as it would make it harder for "unapproved" operating systems to be used in that country.

> the tickets are invalid if you weren't the one driving

I guess what I meant was "the government will punish you unless you can prove someone else was using your device" so you won't be able to escape prosecution by sharing a device and saying "I can't remember who was using it at that time". Similarly, I believe in some jurisdictions a car owner is expected to know who was using their car at any given point in the past so that speeding tickets can be assigned to the correct person.*

Anyway, I can imagine the law going further and matching the dystopian vision of "The Right to Read", which includes this passage: "Of course, if the school ever found out that he had given Lissa his own password, it would be curtains for both of them as students, regardless of what she had used it for. School policy was that any interference with their means of monitoring students' computer use was grounds for disciplinary action. It didn't matter whether you did anything harmful — the offense was making it hard for the administrators to check on you."

* "It is also illegal [in the UK] to decline to provide the driver's details, whether it was you or another person." https://news.jardinemotors.co.uk/how-to/speeding-fine-faqs-w...

◧◩◪◨⬒⬓⬔
101. dane-p+Ap[view] [source] [discussion] 2022-01-09 06:03:26
>>mindsl+hk
> Please drop the hyperbole

What if I had told you 5 years ago that in 2020, people in Western countries would be forbidden from leaving their homes without permission, and would have to show a digital pass on their phone to be allowed to go into shops?

The technology for remote attestation already exists, and it would take less than a year to roll out checks for it across all ISPs in a country. As you say, it would need some sort of crisis for a government to demand it, but an ill-intentioned government with an offensive cyber-war capability could manufacture that crisis tomorrow if it wanted.

We already have authoritarian Western nations like Poland allegedly using cyber-weapons against opposition politicians[0]. I don't think that claiming existing technology could be used in 5 years is a claim that "the sky is falling right now". The main thing holding back such a scheme is that it would force a lot of legitimate users offline, which is why I think 5 years should be enough time to make those affected users a small enough minority that a government could ignore them.

[0] https://www.euronews.com/2022/01/05/polish-watergate-tension...

◧◩◪
127. 66fm47+AJ[view] [source] [discussion] 2022-01-09 10:01:20
>>mrjin+fy
Once the vast majority of devices are remote attestation capable (Windows 11 requiring TPM will accelerate this trend), content providers may refuse to serve you unless you attest that you are running a walled-garden OS that won't allow you to ad-block, capture content, run any sort of proxy server, etc.

At some point, even ISPs might require remote attestation to allow you to connect your device to the internet. The IETF is already working on standards for the attestation of network devices[0][1].

I speculate that there will temporarily (perhaps similarly to iOS jailbreaking, which is not available at this time for the newest devices/iOS version[2]) be exploits allowing you fool the attestation by e.g. redirecting it to another device as the author suggests, but the end effect will be that vast majority of people will be effectively confined to a walled garden and even determined hobbyists will only be able to use their general computation capable devices to access all content (or even connect them to the internet) some of the time.

[0] https://archive.fo/uQULm

[1] https://datatracker.ietf.org/doc/draft-ietf-rats-tpm-based-n...

[2] https://en.wikipedia.org/w/index.php?title=IOS_jailbreaking&...

133. no_tim+jQ[view] [source] 2022-01-09 11:28:11
>>foodst+(OP)
The fact that there is no "user override"[0] feature in any of these security processors is blatantly obvious evidence that they are designed to control and restrict first and foremost. I have read mjg's other posts on the topic and have no reason to believe he is arguing in bad faith but I'm still not convinced one bit.

[0]: https://www.eff.org/wp/trusted-computing-promise-and-risk

◧◩◪◨
137. fsflov+IR[view] [source] [discussion] 2022-01-09 11:45:07
>>messe+od
> > Both Windows and macOS are approaching the iOS security model that depends on manufacturers blessing what software can run on their products, and banning software they don't want users to run.

> That's been said for years, and hasn't held true.

https://news.ycombinator.com/item?id=25074959

◧◩
140. fsflov+lS[view] [source] [discussion] 2022-01-09 11:53:27
>>grawpr+La
> I'm so unbelievably sick of this 'security by corporation, it's what's best for you so accept it bullshit.' I really am.

Then, consider supporting the alternative approaches to security: https://puri.sm/posts/the-future-of-computers-the-neighborho...

◧◩
145. tentac+0V[view] [source] [discussion] 2022-01-09 12:23:16
>>messe+sa
> once FOSS drivers for what is effectively just a new breed of TPM are released.

I genuinely wonder if Microsoft will put any people on this for Linux. They purport to 'love it', but aside from a few Embrace Extend and Extinguish[0] strategies like Edge, WSL, VS Code etc. I haven't seen anything that made me jump out of my chair in amazement.

Maybe they'll surprise me.

[0]: https://en.wikipedia.org/wiki/Embrace,_extend,_and_extinguis...

◧◩◪
164. transp+vb1[view] [source] [discussion] 2022-01-09 14:48:28
>>tentac+0V
Pluton has been supported by Microsoft Linux for several years and their Azure Sphere support contract promises Liunx security updates for 10+ years, https://www.platformsecuritysummit.com/2019/speaker/seay/
◧◩◪◨⬒
187. dane-p+0Z1[view] [source] [discussion] 2022-01-09 19:51:15
>>floatb+521
> How can ISPs do anything close to this when they're not even concerned with how many devices you have?

Unfortunately that's not guaranteed to always be the case. The "Trusted Computer Group" already have ways for network operators to answer "Who and what’s on my network?"[0], and it's possible to set up an IPsec VPN between your device and the ISP where the key is only known to the TPM on your device.[1]

Of course the user could try to proxy requests from an "untrusted" machine to a "trusted" one, and piggyback the connection, but I imagine that applications which allow this won't be allowed in "secure" app stores, and "secure" operating systems would in any case firewall off packets coming from "untrusted" machines in the first place.

[0] https://trustedcomputinggroup.org/work-groups/trusted-networ...

[1] https://wiki.strongswan.org/projects/strongswan/wiki/Trusted...

◧◩◪
188. dane-p+lZ1[view] [source] [discussion] 2022-01-09 19:54:56
>>mkup+N31
As I explain[0] in response to a sibling comment, sadly it won't be enough (eventually) to have just one locked down device on your home network, they will all have to be individually locked down to access the internet.

[0] https://news.ycombinator.com/item?id=29866732

◧◩◪◨⬒⬓⬔
189. fsflov+T12[view] [source] [discussion] 2022-01-09 20:14:50
>>toasta+E41
Consider this: https://puri.sm/products/librem-14. Not plastic. It's 1080p, but for 14" it should be fine for most people. Supports two external 4k screen AFAIK.
◧◩
190. richar+T32[view] [source] [discussion] 2022-01-09 20:31:22
>>userbi+pe
If you are interested about this topic (what OSes should do today) I can advise this talk: https://www.usenix.org/conference/osdi21/presentation/fri-ke...

A bit long but I didn't get bored

◧◩◪◨
195. joseph+ju2[view] [source] [discussion] 2022-01-09 23:42:32
>>heavys+zt2
> you can turn off the TPM via the BIOS

In theory you can. In practice, programs will refuse to run if you do this: https://www.techspot.com/news/91138-valorant-anti-cheat-syst...

That goes for Secure Boot too, btw.

[go to top]