14 ms·
NSA Starts Contributing Low-Level Code to UEFI BIOS Alternative
- mort96 7y agoI hope the Coreboot people are extremely careful about this. Accepting code from a giant agency with the goal of making everyone less safe is very dangerous.
- kbenson 7y agoThe same was said of selinux. I think perhaps the NSA is slightly more complicated with more complex goals than you describe.
- lugg 7y agoAre people really so quick to forget? https://www.miracl.com/press/backdoors-in-nist-elliptic-curves https://www.miracl.com/press/backdoors-in-nist-elliptic-curv... Their goals are at the very least nationalist. As a non US citizen why should I trust them at all? Their main directive errodes my privacy on most levels in most sense of the words.
- kbenson 7y agoAll I said is that they may have complex goals, while pointing out selinux was also contributed primarily by them. I take it that means your position is that they only contribute code for the purpose of exploiting people for national security, and selinux is also suspect?
- mort96 7y agoI mean, being primarily contributed by the NSA is a great reason to not use SELinux, especially if you're outside of the US. AppArmor exists and does a lot of the same job, so why not use the safer option?
- closeparen 7y agoThey're a spy agency, they're presumably capable of not signing their work.
- ENOTTY 7y agoThe chief problem with AppArmor is that its primary mechanism of identifying objects in the system (to which relevant policy rules are applied) relies on paths. In Linux, paths are not usable as strong identifiers because Linux provides a variety of ways to alias those objects. Additionally, some objects that deserve scrutiny from a mandatory access control perspective don't map cleanly to paths.
- ganzuul 7y agoHas that changed? A decade ago when I looked at their charter they were supposed to strengthen security; not weaken it.
- mankeysee 7y agolmao at trusting their "charters" to gauge their real aims
- AmericanChopper 7y agoTheir two missions are Signals Intelligence and Information Assurance. The contributions they make to security technology and standards are going to be equally as effective no matter what country you’re a citizen of. If you think the work they release is motivated by an intention to compromise your security, then you must also believe they are trying to do the same thing to the US federal government. A goal which hardly seems very nationalist.
- lugg 7y agoBoth of those missions are at odds with improving security. Try harder.
- AmericanChopper 7y agoInformation Assurance is not at odds with improving security. It is literally the NSAs mandate to improve the security of the federal government (and other US entities). You’re making an assumption that in order to gather signals intelligence, that they need to trick people into using compromised cryptography. A fact that is not in dispute is that huge portions of the US federal government implement NSA recommendations and standards. For your assumption to be true, the NSA must be intentionally weakening federal systems. Nothing about that seems nationalistic to me.
- cf498 7y agoDid you already forget how long they knew of EternalBlue instead of immediately forcing patches?
- dehrmann 7y agoThe NSA's track record here is mixed, so I'm not sure how to feel. They actually improved the design of DES. > Bruce Schneier observed that "It took the academic community two decades to figure out that the NSA 'tweaks' actually improved the security of DES."
- lugg 7y agoAll they did was make sure breakable-by-them encryption was in use much longer than it should have been. I'd hardly call it a mixed track record. > However, the NSA also ensured that the key size was drastically reduced such that they could break it by brute force attack. https://en.m.wikipedia.org/wiki/Data_Encryption_Standard https://en.m.wikipedia.org/wiki/Data_Encryption_Standard
- mort96 7y agoI got curious and found the post you were referring to (https://www.schneier.com/blog/archives/2004/10/the_legacy_of_d.html https://www.schneier.com/blog/archives/2004/10/the_legacy_of...). They tweaked the algorithm, and that tweak turned out to make it more secure; they also cut the key size by more than half, making sure they could brute force it. That doesn't exactly scream benevolence.
- anfilt 7y agoIt could also be the NSA does not trust proprietary firmware for similar reasons people are working on open source firmware to begin with. The problem is NSA really only knows their motives. The other problem with the NSA is that they have two goals. One is to protect the US and US entities security and the other is subvert foreign entities. If they start publicy recommending coreboot for what they deem senstive installations then their intentions are obviously not malicous. Problem is the NSA may never publicy specify that its for senstive installtions and may only ever be an internal guideline.
- rincebrain 7y agoIt's worth noting that it's even worse than that - some of the NSA employees contributing might not be aware of the motives involved if any are concealed.
- ganzuul 7y agoFrom what I have read of how compartmentalized that organization is; could be no ill intent exists for now. Later however, the knowledge they gain could be used for a different purpose.
- AsyncAwait 7y ago> no ill intent exists for now. Later however, the knowledge they gain The thing is, if there's no ill intent now, there shouldn't be any backdoor to allow them to have any sort of special knowledge later. Unless you mean familiarity with the codebase in general. If they however do insert a backdoor, even without using it for anything now, "just in case", than that in itself is ill intent.
- ganzuul 7y agoI do mean familiarity. Also, there is connections with other people in the project.
- mort96 7y ago
- neilv 7y agoIf there's a concern about that, then maybe some other governments should be paying people to be intimately involved with projects like Coreboot, Linux, GNU, open hardware, etc. -- such that those people are spending enough time that they can understand and keep an eye on things, as well as furthering the projects with their more conventional contributions. (I know Coreboot has had at least a few people who seem to be activists or enthusiasts who are not potentially beholden to anyone, but people getting paid can afford to spend much more time on projects than volunteers can.)
- garmaine 7y agoI won’t be an NSA apologist because in the last 20 years they have in fact made everyone less safe with their undermining of open standards. But for the purpose of accuracy it should be pointed out that one of the two missions of the NSA is to safeguard American commerce, which they did for many decades prior (see: DES, selinux, etc.).
- transpute 7y agoA public STM implementation has long been needed. In 2009, ITL/Qubes wrote about DRTM (Intel TXT, AMD SKINIT) and STM, https://invisiblethingslab.com/resources/bh09dc/Attacking%20Intel%20TXT%20-%20paper.pdf https://invisiblethingslab.com/resources/bh09dc/Attacking%20... & https://www.blackhat.com/presentations/bh-dc-09/Wojtczuk_Rutkowska/BlackHat-DC-09-Rutkowska-Attacking-Intel-TXT-slides.pdf https://www.blackhat.com/presentations/bh-dc-09/Wojtczuk_Rut... > The late launch ... promises to effectively provide all the benefits of a computer restart without actually restarting it. It is hard to overemphasize the potential impact that a technology such as TXT could have on computer security ... We describe a practical attack that is capable of bypassing the TXT's trusted boot process ... As part of the attack we also discuss practical attacks on SMM memory ... Intel's remedy to malicious SMM handler is called STM, which stands for SMM Transfer Monitor. The purpose of STM is to sandbox the existing SMM handler by virtualizing it using VT-x and VT-d technologies. STM should be thought of as of a peer hypervisor to the VMM that is being loaded using late launch. STM is supposed to be measured during the late launch process ... no STM, as of today, is unfortunately available on the market, which yields our attack applicable to all current systems. One aim of our research ... is to stimulate developers to create an STM. The May 2019 version of Windows 10 added support ("SystemGuard") for DRTM-enabled hardware that could benefit from an STM, https://www.microsoft.com/security/blog/2018/04/19/introducing-windows-defender-system-guard-runtime-attestation/ https://www.microsoft.com/security/blog/2018/04/19/introduci... & https://www.platformsecuritysummit.com/2018/references/#systemguard https://www.platformsecuritysummit.com/2018/references/#syst... In 2018, NSA gave a presentation on their STM work, https://www.platformsecuritysummit.com/2018/speaker/myers/ https://www.platformsecuritysummit.com/2018/speaker/myers/ > We describe our work to demonstrate an enhanced SMI transfer monitor (STM) to provide protected execution services on the x86 platform ... Our STM enhancements create a protected execution capability by extending the STM to support additional VMs (PE/VM) From a coreboot developer, https://twitter.com/_zaolin_/status/1055474061428572162?s=21 https://twitter.com/_zaolin_/status/1055474061428572162?s=21 > We are currently implementing @intel #TXT and #SRTM measured boot support as part of Google's verified boot which can be used on all supported platforms in @coreboot_org
- ackbar03 7y agoTerrific! About time we had some people with proper technical expertise contributing to our everyday systems. #Taxpayer dollars put to use
- neilv 7y agoI applaud moving away from UEFI, to something simpler and more transparent, and something not hostile from the start to open source. I like how relatively simple my Coreboot systems are. The sooner that Coreboot can minimally initialize hardware and invoke an open source bootloader or Linux kernel payload directly, the better. (Bonus, if the kernel doesn't need legacy PC BIOS services.) Regarding concerns raised in the article, I'd be more suspicious of closed microcode and firmware blob parts, than of open source parts. Many knowledgeable eyes on the open parts can help. Though, of course, open parts could, in theory, be crafted to provide affordances for naughtiness involving closed parts. I have mixed feelings about recent initiatives to push closed updates more widely and reliably, and look forward to more-open hardware.
- maze-le 7y ago>> Many knowledgeable eyes on the open parts can help Well, that is the problem. No one (well, say very very few) do code reviews on their own -- they do not provide as much benefits as coding (learning and reputation wise). Its a bit like the problem in academic publishing: Replication trials are not prestigious enough, so they are done far to rarely, albeit they are very important. There are companies out there, that will do it for money, and many great open source products have been reviewed that way. But I don't think this happens often enough.
- PakG1 7y agoWhile that's a valid concern, the real question should be: despite that, is it more trustworthy than closed source or not?
- maxheadroom 7y ago>...they do not provide as much benefits as coding (learning and reputation wise). I think that's because code reviews have changed from being helpful form of preventing you from stubbing yourself on your toe to, "Why would you do this...? No0b..." kind of format. If code reviews were weighted as equally beneficial to the code and as being approached as being beneficial to the growth of the developer making the change, then I could see it having much more benefit in learning; however, that would take a dynamic change in the general culture. ...but saying that code reviews don't provide as much learning benefit is to just infer that there's nothing that can be done about that when, in fact, there is. :)
- jchw 7y agoOh, cool. I really enjoy coreboot in principle, but I wish more systems supported it. None of my daily drivers are running Coreboot, as of today. I’m pretty inept when it comes to low level stuff, but I’ve been playing around trying to port Coreboot to an unsupported laptop board. It’s probably futile, but it was pretty exciting seeing some serial output for the first time. From lurking around, one of the more surprising things I’ve found is that DDR RAM initialization seems to be the single most difficult aspect of the whole boot process, or at least on typical PC platforms. Not to say everything else isn’t also difficult; the debugging tools available to the general public for firmware are fairy rudimentary.
- incompatible 7y agoThere was a story a few years ago about "Intel Boot Guard" making it impossible to install things like Coreboot. I don't know if that was ever resolved.
- ENOTTY 7y agoThat's not strictly true. In my view, Boot Guard is sort of like a locked bootloader. In Boot Guard, the CPU verifies a signature on system firmware before loading it. The signature is verified using a public key from the platform manufacturer. So under the Boot Guard regime, the platform manufacturer essentially gets a vote in whether your platform can run modified firmware level code. Or you find a jailbreak.
- incompatible 7y agoOK, in that case I'd be very interested to know which motherboard manufacturers are friendly to installation of replacement firmware and which not. coreboot.org links to libreboot.org, but the list of supported hardware is pretty short: https://libreboot.org/docs/hardware/ https://libreboot.org/docs/hardware/
- ENOTTY 7y agoI think purism might be amenable. Worth double checking with them.
- stakhanov 7y agoWhen I read the headline, my brain first parsed it as a scandal breaking, then I read the article and found that it was not that. I just hope the open source community will be very serious about auditing that code and not accept any blobs coming from people affiliated with the NSA.
- unnouinceput 7y agoYour brain was correct...something like 5 years before correct. I bet in 5 years this will be a scandal breaking because whatever NSA code gets adopted, it will be unveiled later as a backdoor.
- kryptiskt 7y agoIf the NSA wanted to add a backdoor, they would likely not do it in a patch openly contributed by them.
- stakhanov 7y agoIt's probably more subtle than that: coreboot is a hodgepodge of open source aspects and oblique blobs that contain god knows what. That's why libreboot exists (shares a lot of coreboot code but doesn't allow blobs). If the NSA wants to do some dirt, then the dirt will probably go into the blobs and/or happen as a result of how the open source code interacts with the blobs in ways that are not obvious to anyone who doesn't know what's in the blobs. So I can see how getting some influence in the coreboot community would be of strategic value to them to implement schemes to weaken IT security for whoever they want to weaken it for.
- DigitalTerminal 7y agoLOL! Trusting NSA for UEFI code is like.... I hope cryptographers and hackers better than what NSA has go through that code with a fine toothed comb.
- ShinTakuya 7y agoThe title is pretty shitty but you should at least click through before making a comment. This is a contribution to Coreboot, which is an alternative to UEFI/BIOS. As much as I dislike US government agencies in general, I think this time they have good motive to provide good code. Undermining this code would also undermine the systems of other government agencies. In this case, I trust the NSA more than random contributors on the internet that have less known motives and may in many cases be agents of foreign spy agencies.
- maxheadroom 7y ago>The title is pretty shitty but you should at least click through before making a comment. Something, something, site guidelines about suggesting the commentor didn't RTFA. >I trust the NSA more than random contributors on the internet that have less known motives and may in many cases be agents of foreign spy agencies. To be fair, I think that you stated the OC's point and failed to recognise it: Your domestic spy agency is another's foregin spy agency. Why should they trust yours over theirs? Both have (presumably) equally less-known motives, yeah?
- cannedslime 7y agoFor non-americans, NSA is a "foreign spy agency"...
- p_l 7y agoAnd "illegal combatant"
- p_l 7y agoCoreboot is NOT an UEFI alternative. Unless you want to regress computing to closed platforms (it can be open source and closed platform).
- arianvanp 7y agoI applaud this and I think the coreboot team is good enough to be critical and diligent about patches they will receive.
- unnouinceput 7y agoI wouldn't trust NSA with a "hello world" program. The way they might phrase it could trigger hidden feature in linker which in turn could trigger hidden instructions in CPU making the 3 lines program a backdoor into your system.
- indalo 7y agobrings the concerns that seem to have been forgotten about the strange CPU behavior with their preffered version with elliptical curve
- steve19 7y agoYou better avoid the Linux kernel then as the NSA have contributed a lot more than 3 lines.
- asveikau 7y agoI hate to break it to you, but NSA's hello world program is line by line identical with the others you have seen. How did they infiltrate us so badly! It's a scary day for anyone running mission critical hello world programs in production.
- macpete 7y agotimeo danaos et dona ferentes
- vinay_ys 7y agoIt is the job of journalists to ask these controversial questions about NSA's role as a saboteur vs contributing in good faith to really improve the security of systems. So, nothing wrong with that. Healthy skepticism and all that. In my experience, when closed source research teams start to contribute to well-run open-source projects there can be valuable contributions. But it highly depends on having independent and technically competent maintainers with strong personalities who are not easily manipulated into accepting patches that they don't understand or violate their technical principles.
- Stay_frostJebel 7y agoI hate it when news outlets assume readers don't know what they are talking about and use analogies like "UEFI BIOS Alternative" without mentioning the actual thing on the title ("Coreboot").
- caymanjim 7y agoThis isn't a bad thing. I had no idea what coreboot was and wouldn't have clicked on the link. They're appealing to a broader audience without being clickbaity.
- josteink 7y ago> I hate it when news outlets assume readers don't know what they are talking about How about don’t know what they’re writing about? I sure as hell have never heard about “Windows UEFI” before.
- floatboth 7y agoEspecially when it's flat out wrong. Coreboot is low level init code that launches a payload. Which can be TianoCore EDK2 or SeaBIOS.
- bbcfake 7y agoLower level to "stop" Russians ?? :))))
- achingtooth 7y agoPeople in this thread are talking about how they wouldn't trust the NSA at all. I went to a presentation and talked with people from the NSA before and at face value they seemed like a silicon valley tech company. In their presentation they talked about how they were interested in open source, diversity, big data, artificial intelligence, and all the other buzzwords. They all seemed like they genuinely thought what they were doing was helping people. I know what they've done (and continue to do) but it's strange being able to attach a face to an action. You're more likely to believe them and buy what they are saying. I suppose the best thing to do is check over their code and accept it if everything looks good. They probably are being genuine. As a an extra piece of information that I found interesting, they were pushing the diversity stuff hard. Everyone that gave the presentation were women (and they weren't low level people), they had an African-American person that worked there talk about how inclusive it was, they talked about how they're super accepting of LGBTQ+ people, and on and on. The tech stuff was for like 5 minutes, then the rest was on diversity (at a tech presentation, looking for recruits). I'm not exaggerating.
- stakhanov 7y agoDon't drink the kool-aid. ...they probably have a bunch of LGBT African-American women whose sole job it is to be visible. The NSA is one of the few companies legally allowed to do stuff like only hire you if you're a U.S. citizen and even say so in their job advertisements, and keep you out of certain kinds of roles (the ones where the real action is, probably), without being held to transparency standards and nondiscrimination laws that would apply to private corporates. They can always cite undisclosed nonspecific security concerns rather than having to say "We didn't allow that person into that role because it isn't a middle-aged white guy".
- achingtooth 7y agoI'm 19. I've never been a fan of the NSA, my bad I should have been more clear with what I meant. At least I got a cool notepad https://i.imgur.com/yu103Lg.jpg https://i.imgur.com/yu103Lg.jpg
- 7y ago
- zeristor 7y agoI was setting up Windows servers 15 years ago, the company was chuffed to be using special secure NSA tweaked versions of the OS, which was nice of them...
- NelsonMinar 7y agoCould you imagine the shitstorm if Coreboot was accepting code from a Chinese intelligence agency? I'm American myself and am glad NSA is on my side. But they've proven themselves time and time again to be completely untrustworthy when it comes to securing our computer systems.
- oil25 7y agoThis is great news - not only will competent NSA hackers improve Coreboot, but their participation will usher in even more scrutiny and review of the code. This is a win for everyone.
- throwawayf987df 7y agoIf Govt's around the world want to really secure their IT systems, then just remove the laws that get hackers into trouble for hacking Govt systems. Still keep the law for non-Govt systems or for changing anything on a govt system and then the Govt can sit back and enjoy free pen testing.
- turk73 7y agoDoes anyone trust the NSA or the CIA at this point? I mean, it has been damn near transparent that these organizations have been up to no good. Maybe we the people should get tougher on these rogue agencies? Crickets. You're only afraid because you think they can crush you, and you're right--they can crush you. But they can't crush us all.
- peter_d_sherman 7y agoIn reading the discussion here, I'd like to bring up an error that the News Media typically makes when discussing large groups of people... That is, they implicitly or explicitly judge large groups of people -- most notably other countries / other nationalities -- by the actions of the few. Applied to a group, in this case the NSA, I don't believe that they should be judged by the actions of a few of their members... I am sure there are both socially positive and socially negative actors in the group (and everybody in between), thus it is disingenuous to judge the group in either direction... Me, I'm completely neutral about the NSA, with one exception, and that is that I feel that companies should NEVER be exposed to secret NSL's... either make those communications a part of the public record for congressional and other legal/legislative oversight, or don't send them in the first place! But that's not the people of the NSA, whom I hold harmless... That's part of the mission of the NSA, and well, mere mortal citizens are probably not going to change that anytime soon... If anyone really wants a secure computer, build yourself a VAX-11/780 out of transistors, and write the operating system too... (If I ever did, I'd put on a few of my super-secret cooking recipes ("Mmm, however did you get those BBQ ribs to taste so good? It's a secret!") along with a note "If you got this message, you are definitely elite, and please don't delete my BBQ sauce recipe!"). Try and get that with no RF link in the electronics... I double dare you...<g> In the meantime, back up your files, and audit your communications regularly for anything you wouldn't want on the news 24/7... <g> But the NSA? Not evil...
- crankylinuxuser 7y ago>Try and get that with no RF link in the electronics... I double dare you...<g> Build my GPL3'd Signals Intelligence device, and you can audit it :) https://www.rtl-sdr.com/the-radioinstigator-a-150-signals-intelligence-platform-consisting-of-a-raspberry-pi-rpitx-2-4-ghz-crazyradio-and-an-rtl-sdr/ https://www.rtl-sdr.com/the-radioinstigator-a-150-signals-in...
- peter_d_sherman 7y agoDidn't know that that existed... Looks really nice!!! (Although, technically, to be really secure, you'd have to audit the entire RF spectrum AND lower frequency wavelengths, i.e., ultrasonic AND have to figure out what to do with signals that are intermittent and/or frequency hop!!! Then of course you have the power supply lines and the ability to send super-low frequency signals via all of that... In other words, you'd have to audit all wave frequencies from all connected devices, simultaneously, and keep in mind such possibilities as frequency hopping, and intermittent periods of silence... like if someone really wanted to be stealthy, they could send something like 1 byte per hour... and that hour is randomized, so it's actually a random interval between 50 and 70 minutes... and/or hide that in random radio noise... the possibilities are endless... <g> (This is why I leave security to the security people and just assume that no information is private anymore...)) But, all of that being said, your link looks really nice, and I didn't know that existed before! It looks cool and worthy of experimentation!
- stjohnswarts 7y agoWhat could possibly go wrong.
- losttheplot 7y agoKeep proven bad actors and miscreants, be it government or large tech, out of establishing standards.