Hyper-Threading, Linus Torvalds vs. Colin Percival 396
OutsideIn writes "The recent Hyper-Threading vulnerability
announcement has generated a fair amount of discussion since it was released. KernelTrap
has an interesting article quoting Linux creator Linus Torvalds who recently compared the vulnerability to similar issues with early SMP
and direct-mapped caches suggesting, "it doesn't seem all that worrying in real life." Colin Percival,
who published a recent paper on the vulnerability,
strongly disagreed with Linus' assessment saying, "it is at
times like this that Linux really suffers from having a single dictator in charge; when Linus doesn't understand a problem,
he won't fix it, even if all the cryptographers in the world are standing against him.""
He won't fix it? (Score:5, Insightful)
Re:He won't fix it? (Score:3, Insightful)
Re:He won't fix it? (Score:5, Interesting)
He doesn't say "I don't want a fix for this anywhere in the kernel" anywhere. Just that he doesn't think it's a very critical issue.
If someone else does the patch for him, why wouldn't he accept it?
Re:He won't fix it? (Score:5, Interesting)
What this bug amounts to is this: When a program is performing calculations using secret data like an RSA key, it is important that the data access patterns do not depend on the secret data, since these patterns can be analyzed by an attacker.
An example of a classical vulnerability of this sort is using the c function strcmp to compare the real and the supplied password. By timing multiple runs you can get a decent estimate of how long time the strcmp function took, which means you can guess which character was first differing character in the password.
The security flaw in HT is that a process running on a HT CPU can get quite a lot of information about the data access patterns of the process on the other virtual CPU on the same chip. In other words, the severity of any library bugs which cause different access patterns on different secret data has been severly increased.
Re:He won't fix it? (Score:5, Insightful)
Can I buy some pot from you?
Maybe that would work with a ONE MEGAHERTZ PROCESSOR. But do you have any idea how fast processors are these days? And how likely any deviance in the cache state, IO controller state, page faults, multi-user latency, or power management will throw your precious timings right out the window?!?!?!
I mean, c'mon, think about things before you say them. Even REAL TIME SYSTEMS AT NASA don't run with enough consistency to be able to tell WHICH CHARACTER IN A STRCMP OPERATIONS fails.
Re:He won't fix it? (Score:5, Interesting)
Secondly, just because the clock is ticking fast doesn't mean you can't count them as they go by. The same process run at 1Mhz and 1Ghz is going to look substantially the same, because you're monitoring them faster too. If anything, you're less likely to get interrupted by I/O during the process.
Re:He won't fix it? (Score:5, Interesting)
It's true that there are various "noisy" and uncontrollable aspects of modern CPUs, but if you take enough measurements and can establish a control, you can get valid information. It will not give you "the answer", it may help you to eliminate a set of data that is "not the answer". That is extremely valuable.
Some people actually monitor chip power supply variations to try to get this information. I won't pretend to understand how they get meaningful data from this, but it involves knowing the system under attack very well, and wanting the data really, really badly. Hence very high security systems shield their processors magnetically and inject "noisy" (bogus, unrelated) operations to cover this. It sounds like a pipe dream, but I've visited places where these systems are built (think Blue), and read some of the papers describing these exploits.
His point is valid, ANY information an attacker can get significantly reduces the integrity of the encryption. I just suspect Linus' point is that most (I said most, don't flame me) linux server deployments are geared towards not letting people on the server in the first place. This vulnerability seems to be valuable mostly to people who already have access to the system.
Re:He won't fix it? (Score:5, Informative)
Maybe the original poster was referring to the DEC-10 page fault password insecutity that was based on strcmp returing as soon as it encountered once wrong character, based roughly on the following idea:
1) Place Password with 1st/2nd character on a page boundry.
2) Clear cache
3) Call Password check.
4) If no page fault occurred, then the 1st char must be wrong, change it and goto 3
5) if page fault occurred, 1st character is correct (as 2nd char was checked), move password so 2/3 char is on page boundry and repeat.
In this way, you can reduce the attack by a huge amount, for a n length password the brute force attack needed goes down from 256^n to 256*n.
So, yes, attacks based on which character in strcmp fails have worked in the past, so it is valid to try and not make the same type of mistake again!
A possible solution (Score:5, Informative)
Here's another option: since this vulnerability depends on using the L1/L2 cache states to ferret out information, remove that from consideration. When processing an RSA (etc) key, have the code temporarily lock out context switches, AND have it take a fixed period of time to compute the result (or portion of the result), followed by flushing the cache. No data is left in the cache to analyze. The execution time of the code is fixed, so no dat leaks there. You don't have to make the algorithm a constant-time calculation if you can pad it at the end to the maximum (or close enough to the theoretical maximum that there's no true information leakage in practice). This helps avoid the potentially very slow algorithms that are truely constant-time. And flushing the cache at the end of each (portion of the) calculation removes that as a leak. (You need to flush it before waiting for the allocated time to elapse, note, and include max flush time in your calculation, and if possible factor into your calculation possible interrupt effects.)
A pain? yes. Requires extensive mods to crypto algorithm implementation? Yes (though perhaps not to the core calculation.) Requires OS support? Almost certainly. Requires HW support? Would be helpful but probably not required. Loss of performance? Yes, though far lower than disabling HT I imagine in normal cases (when not decrypting/encrypting).
Also, some of the restrictions above can probably be eased if the crypto algorithm is carefully designed and matched to the hardware.
Disclaimer: I am NOT a crypto geek! I have worked in processor and cache design, though.
Re:strcmp vulnerability. (Score:5, Interesting)
Re:strcmp vulnerability. (Score:3, Insightful)
Re:strcmp vulnerability. (Score:4, Informative)
The normal solution is to store both the the real password and the password supplied by the user in a memory area that is as large as the maximum password length and zero padded and do something like this:
Note that it is also vital thet the password supplied by the user is zero padded and the calculation time can not depend on how long that password is. If it did, it opens up to man in the middle attacks. Evil, huh?
Re:strcmp vulnerability. (Score:3, Interesting)
Algorithms like SHA work on blocks of data, so what you need to do is make sure the code for padding the password to full block length doesn't change the runtime.
Re:strcmp vulnerability. (Score:4, Interesting)
strcmp may be fast, virtual memory is not. You just make sure that your password ends up split between two pages in the comparing program - for example by padding the previous thing you send over network with spaces. Then generate some activity to make sure most of the things are paged out. Now if the part of the password on the first page is correct, the program will take longer to return failure and an extra page should be visible through vmstat. By repeating this with different alignment, someone can guess your password character-by-character.
I am afraid multi-user systems are there to keep students from stealing each other's homework, not for protecting military secrets. Over the Internet, such small variations in timing should be impossible to measure. You can insert a random delay every time your program processes an incorrect password for good measure.
Re:He won't fix it? (Score:5, Funny)
If I remember correctly, there hasn't been a shown exploit for this yet. It's better to wait and see before fixing something that may not matter later.
Re:He won't fix it? (Score:3, Insightful)
Re:He won't fix it? (Score:5, Insightful)
So MS$ shouldn't fix problems in IE until an exploit has been shown for it?
It's better to wait and see before fixing something that may not matter later.
Its better to just fix it and be safe than wait and see if something happens later. It may not be top priority, but remember this "wait and see" approach to security is exactly what got MS$ into so much trouble with users. We don't need the same for Linux.
Bzzt! Back the paradigm up here. (Score:5, Insightful)
Potential performance problems are things you should defer on until proper profiling can be done (unless they're total show stoppers). Security and correctness are things you cannot ignore except in extreme cases. Security is particularly important to nail down, because it can result in your customers losing data (even data not pertaining to your app), which is the first no-no of software.
Application software has four priorities, in this order:
- Safety (shouldn't destroy data)
- Correctness (do what it says it does)
- Security (don't do anything else)
- Performance (do it fast)
YMMV, of course, sometimes correctness falls below security, and occasionally performance goes above correctness in some mathmatical functions (if doing it correctly would take a decade and doing a close approximation would take a day, obviously you want the approximation and then a heuristic). In this case, I'd say proper fix is to disable hyperthreading by default, and make sure the user is aware of the hardware bug/consequence of using HT when they decide to turn it on. You need to let the user decide if they're willing to accept the security risk or not.The Linux Kernel Developers may decide otherwise, but that's how I'd call it if it was in my shop. It's a hardware problem and the software fix is not obvious.
Re:Bzzt! Back the paradigm up here. (Score:3, Informative)
In my experience, correctness is caught here, but security tends to slip through the cracks. It can be very hard to unit test for security (although, a well tested system tends to have fewer buffer overflows and whatnot).
Re:He won't fix it? (Score:5, Insightful)
Re:He won't fix it? (Score:4, Funny)
Re:He won't fix it? (Score:5, Funny)
Re:He won't fix it? (Score:3, Interesting)
Programmers hell, most mechanics I know have this attitude. 'Well, it stopped, didn't it. We'll fix it later'
I wouldn't blame software developers for this attitude, it seems to be common for humans (or at least Americans) as a
Re:He won't fix it? (Score:5, Interesting)
Re:He won't fix it? (Score:3, Insightful)
Re:He won't fix it? (Score:2)
now..
what are the real threats from the vulnurability?
the author of the paper seems to see there to be some crapload of problems from this, yet I don't quite see it so(the attack route being quite far fetched imho). so much that a lot of other people probably saw this same 'problem' earlier as well but didn't really think of it as anything.
AND
is it something that can be done something to in the kernel and not in the suppoesdly vulnurable openssl? like, it is a 'feature' rather than a bug.
Re:He won't fix it? (Score:3, Insightful)
Some people are just keep pushing their flawed agendas.
Disclaimer: i did read the whole thread.
Re:He won't fix it? (Score:2, Insightful)
The only fix that kinda works is disabling hyperthreading. But on a dual core processor the problem is there as well (if there is a shared cache somewhere). And switching of the second core because of that would be stupid.
The problem Colin points out (getting an RSA key) is a userland problem anyway, so there is nothing to fix for Linus... fixing cache eviction covert channels in the kernel is possible, but not without losing a lot of performance.
--Blerik
Re:He won't fix it? (Score:3, Insightful)
I think pretty much every distro in existence maintains its own patchset for the vanilla kernel (effectively maintain their own "fork", I guess), and it is not causing any problems. In brief, if Linus does not accept a useful patch (or essential!), the distros will, and so this "forking", if you can really call it that, will have no neg
Re:Not Accurate (Score:2)
Re: (Score:2)
Re:He won't fix it? (Score:5, Insightful)
Re:He won't fix it? (Score:5, Interesting)
The best solution is to just fix the crypto libraries as a short term solution, and for Intel to fix the chip in future iterations as a long term solution. Mucking about in the kernel and having other unknown effects seems like a bad approach when the problem can be fixed elsewhere.
Re:He won't fix it? (Score:3, Insightful)
Fixing the crypto libraries is actually a non-trivial task. There are many of them and ensuring there is no information leakage is very difficult to achieve.
Fixing the chip may well be impossible -- it sounds to me like the only way to prevent this kind of thing from being a problem is to give each thread its own instruction fetch pipeline, dedicated
Dictator? (Score:5, Insightful)
Come on.
Comment removed (Score:5, Informative)
Vendor kernels contain patches (Score:3, Insightful)
This is one of the good aspects of open-source software: If you disagree, you can fork or simply distribute a patched product.
Re:Dictator? (Score:5, Funny)
He doesn't administer a reign of terror, just the occasional light shower.
Re:Dictator? (Score:2)
As much as I hate to say it, the truth of the matter is this: Linux essentially belongs to everyone in that we can all make our own forks if we feel it necessary. Linus simply directs the pack of juiced up monkeys in the development, but if enough people lost faith in him, a new fork would
Single Dictator? (Score:3, Insightful)
If Microtoft decides not to fix it, then no one else can.
so which one is a single dicatorship?
Re:Single Dictator? (Score:2, Interesting)
Re:Single Dictator? (Score:3, Informative)
Re:Single Dictator? (Score:3, Insightful)
Most dictators have tweaked the kernels to suit their needs anyway, nothing stops them from implementing this in their own kernel version.
In the vanilla tree yes (Score:2)
bad tactics from Colin Percival (Score:3, Insightful)
If he and "all the world's cryptographers" can't do that, then Linus' pragmatism beats the cryptographers paranoia (their defining quality, in my experience) into a cocked hat.
If an exploit can't actually be exploited, it's not and exploit.
Re:bad tactics from Colin Percival (Score:4, Insightful)
The point of this debatte is that the Intel implementation sucks, it allows you to spy a lot on processes running on the virtual CPU. Sure, there are better alternatives than disabling HTT like the suggestions of Colin to only schedule threads of the same program on the virtual CPU. Actually, that is something you want to do anyway or otherwise you can seriously loose speed and drop under the performance of a processor with HTT disabled.
Speaking of paranoia, it is often not a bad thing to have, many big security problems can be avoided. Oh, I forget to patch the Linux box next door.
Re:bad tactics from Colin Percival (Score:3, Insightful)
Being able to read arbitrary memory from another process is a big security flaw, as illustrated by the Minesweeper Hacking [codeproject.com] sample. But for a kernel programmer it's a minor deal for server security as it needs a local/remote exploit to run code on your box. Even then it is a readonly exploit, which decreases exploitability unless we're talking about stuff like SSL certs or GP
Other examples (Score:3, Insightful)
If a process is leaking information somewhere, then there will be people clever enough to pull that information out.
That said, it seems that this is more of a library problem
OK Colin, Well done (Score:4, Insightful)
Re:OK Colin, Well done (Score:3, Insightful)
Re:OK Colin, Well done (Score:5, Insightful)
Re:OK Colin, Well done (Score:2)
Re:OK Colin, Well done (Score:3, Interesting)
Except that for MS problems you'd probably not actually hear about it unless an exploit has been found (MS would of course keep quiet about it, and others would probably not find out other than by creating an exploit).
However, IANASE
Linus the great Dictator (Score:3, Insightful)
Colin needs to cool down a bit. At least the Linux distros (SuSE, Red Hat, etc are the ones which will get a problem from affected systems) are going to get patches for it. From Linus or any other Kernel developer.
Open Source, YOU win! (Score:2, Informative)
Linus and RMS (Score:3, Funny)
Re:Linus and RMS (Score:5, Interesting)
but... but... but... (Score:4, Funny)
Re:but... but... but... (Score:4, Funny)
Re:but... but... but... (Score:2)
Re:but... but... but... (Score:2, Funny)
I want a keyboard man!
Micro-Managing Minutiae Mayhem (Score:3, Funny)
This is interesting logic: The idea that the creator of an organization must understand minutiae and micro-manage everything that the organization does.
Interesting indeed...too bad it's fallacious. (Although it might explain what is taking Longhorn so long to come out - I can see Bill Gates searching Google for whitepapers on file systems, search algorithms, GUI's, etc.)
Re:Micro-Managing Minutiae Mayhem (Score:3, Funny)
At least Linus.... (Score:2, Informative)
From the original article [daemonology.net]
Re:At least Linus.... (Score:5, Insightful)
Re:At least Linus.... (Score:2)
It does go a long way to show the implied argument that the topic is important because the person is "obesssing" about it, is invalid too.
But I guess that wasn't the point of the interview quote in the first place.
Re: (Score:2)
Comment removed (Score:4, Informative)
Fixing is easier said than done (Score:5, Insightful)
Fixing is easier done than said: WBINVD (Score:4, Informative)
Re:Fixing is easier said than done (Score:3, Insightful)
Sounds like it doesn't fix the problem, it replaces it with another (reduced performance). That's not a *real* fix to any problem.
Compare: doing suicide to "fix" personal problems.
Paper author just wants attention. (Score:4, Informative)
All said cyrptographers should buy a non hyperthreaded cpu, or turn it off.
I mean if you use GPG [gnupg.org] on most machines, it will issue you a warning about Insecure Memory. That is someone could potentially harvest data from disused pages in memory.
These cryptographers would use a secure memory system. I'm happy hoping that MI6 isn't running a remote memory exploit on my box.
Re:Paper author just wants attention. (Score:2, Informative)
Re:Paper author just wants attention. (Score:3, Informative)
So even if you did not have any spyware running when you encrypted your secret files on where is the key to satellite that could melt the poles and flood all coastal cities, you would still be in trouble when 007 gets your laptop and searches throught your swap partition and find a suitable random looking bytes that luckily can decrypt everything, once more the free world is sav
This isn't just about stealing crypto keys. (Score:2)
Stuff like timing attacks. A timing attack that might have been difficult or impossible before, may be possible or trivial now.
No crypto involved.
This sort of attitude is pretty common (Score:5, Insightful)
As someone pointed out, yay for linux being free. As one or two above pointed out, someone who does care with the knowledge will write a patch. It'll get implemented as an option in the code, and if shown to be unobtrusive enough, may even get turned on by default.
Re:This...attitude is pretty common - Thankfully! (Score:2)
From this it should follow that we listen to them, not that we dismiss what they say!
I'm glad that the construction engineers who design the bridges we ride over "study [bridges] a lot", and aviation engineers who designed the plane I am about to ride on "study [planes] a lot" and you bet they "have a vested interest" in safety.
Praise the experts who "study a lot", for without them, we'd all be dead. :-)
Re:This sort of attitude is pretty common (Score:4, Interesting)
Well, no, however it is important to many people.
I, like many others here, run an e-commerce web site on Linux servers. In my case, my servers are virtual servers shared with other users who I do not have any knowledge of at all.
If the server used HT, it would be possible for one of those other users to run an exploit on the server to crack my e-commerce site's private key. Fortunately it doesn't, but my ISP could upgrade at any time...
Hence, this is an issue that effects me and my customers, and I seriously hope that a fix finds itself into either apache mod_ssl or the mainline Linux kernel PDQ.
Re:This sort of attitude is pretty common (Score:5, Insightful)
That's really what's up for debate here. Whether the patch should be in the kernel-land or in the code user-space (mod_ssl, for your example).
The only realistic patch you could do in kernel-land is to simply disable HyperThreading. This works, but seems like a poor way to go. Any other form of patch in kernel-land just makes the attack harder and thus doesn't really work or it degrades performance way too much to be practical.
But fixing it in userspace is somewhat easier to do, albeit you'd have to fix *every* user-space program that's susceptible to this sort of thing.
Let's talk about the problem in general terms. When a program is doing some kind of computational stuff on something you want to remain secret, then it has to make some assumptions. Assumptions like the hardware is secure, or that it's not running on a virtual machine that's recording everything it does.. That sort of thing. You can come up with all kinds of ways to crack it like an egg if you work outside the box a bit and have total control of the machine it runs on.
This problem is attacking one of those assumptions, namely that another process can't time the secret computations accurately enough to perform a timing attack. With HT, you have two things running on the same core, and so it is somewhat easier to do this sort of attack.
So userspace programs that do secure computations have had one of their assumptions broken by HT. To remedy it, they need to rethink their assumptions. They need to or ensure that they perform equal timings regardless of the computations being done and so on. This is not particularly simple, but it's probably not particularly hard either.
Of course, the attack is still largely theoretical. All it's been shown is that it's "possible", not that it's "easy" or even that it is indeed "doable". For one thing, without having some kind of clue as to the algorithim involved or some idea of what to look for, all you get are a bunch of timings. You still need to do some things to trigger it at the right time and in the right way as to be able to derive information from this channel.
But crypto guys are paranoid like nobody else, and so they're naturally worried about this sort of thing. Mainly it's worrying to them because it's not a mathematical attack, which they're more used to. Modern crypto works based on theory and algorithims and such, and the idea that the algorithim being correct (for a given value of "correct") isn't enough to protect the security of the data is extremely worrying. A real world implementation of these algorithims now has to take some more real world facts into account, and this bothers them, of course.
Linus is basically right here. The kernel is simply the wrong place to fix this. It doesn't ensure that processes cannot spy on other processes via subchannels like this, nor should it. If you're paranoid enough to think this is a real thing to guard against, then your secure code should take it into account. Existing code doesn't do that, and would need to be changed *even* if the kernel was patched. Because how do you know that your kernel has been patched? How do you know that you're not running on an HT processor? You can't know for sure, so you simply assume you are and take steps to make timing attacks fail. Because if you don't, you can't reasonably say that you've attempted to secure the code in this way.
Re:This sort of attitude is pretty common (Score:3, Insightful)
It may be possible, yes. But plausible?
this is an issue that effects me and my customers
It MAY affect you and your customers at some later time, but right now it doesn't.
If you're THAT concerned about this issue, I assume you're going to call up your ISP and transition your site onto dedicated machines? Isn't it worth the extra cost to be assured that so
Another Fairy Tale... (Score:4, Insightful)
Scene: A wispy cloud scuds across the sunny blue sky. Not much happening, and the cloud is hardly even black.
Chicken Little: The sky is falling! The Sky is falling!
The Penguin DictatorNo, not really. It might fall, but it's very, very unlikely. So calm down!
Chicken Little: I strongly disagree. The sky is falling! And because you do not understand the problem we're all going to die!
The Penguin Dictator:Listen here. It's almost certainly not going to fall, and I need to worry about real problems!
Chicken Little: (Runs screaming to the nearest coffeehouse with free wireless, where he types incessently:) The sky is falling! The Sky is falling! Tell Slashdot! Tell Tom's Hardware! Tell Cnet! Tell Linux Business News!
The Penguin Dictator: Sigh. (And then he gets back to work. He looks up at the audience) They just do not get it, do they?
The Windows Dark Lord: (Rubs hands together) Excellent, MOST excellent. (Yelling) Bring me my marketing minion!
Marketing Minion: (being drug in by a bald guy yelling at him) Yes, O Master!?
The Windows Dark Lord:Tell all the peasants that the sky is raining huge chunks of fire and dung! Tell everyone, tell them now! And have our independent consultants work on this day and night, night and day! Make sure that they independently tell everyone that they can easily avoid falminf chunks of sky dung if they stand behind our Windows! And RAISE the price!
Some Guy At Some House In Some City Somewhere: "Wow, that was easy. Let me send this up to the Penguin Dictator. No sky ever fell, and that cloud is easily blown away. Nothing happening here, move along."
The Penguin Dictator "Well that was easy. Include this patch in the next day's weather update!" Marketing Minion: Press Release!!! Millions killed by falling flaming sky chunks of burning dung with brain eating worms who eat children!!! Run for your lives!!!!
Laura Didio, munching a do-nut"If you would hide behind Windows, the sky would stop falling! Your children would be safer and the world a better place." (looks at stoick ticker, says to self) 'Excellent. MOST excellent. Bring me a donut!'
The Penguin Dictator "Sigh. Why didn't I just keep Sky 0.7a for myself? Why the bother, wy the bother?"
EPILOGUE: No one was ever hurt by the piece of sky that never fell, and Chicken Little kept looking upward for another cloud to rant about.
The End.
Re:Another Fairy Tale... (Score:4, Insightful)
1. Microsoft is just as potentially vulnerable as Linux. Their dirty laundry just doesn't get aired in public.
2. The fix is non-trivial and non-obvious. It's not a simple buffer overflow. Any patches are likely to have serious repercussions on kernel performance. e.g. disabled HT, ensure only two threads of the same U.I.D. are scheduled on the same processor, flush cache at every context switch, etc. It looks that Linus is unwilling to accept them unless this vulnerability moves more from the theoretical to proven.
Great (Score:2, Insightful)
Yeah, just like a mouse driver having full access to kernel security structures and raw disk partitions, it doesn't seem all that worrying at all (except when your entire system crashes thanks to a buggy sound driver, or you get rooted, or...).
Not fixing this design mistake while laughing at respected experts in the field reminds me something [google.com]. I was hoping that we as a community might have became a little bit more mature during the last decade, but I seem
Re:Great (Score:3, Insightful)
I don't think Linus was "laughing at respected security experts" at all. For one thing this guy isn't a respected security expert - in my experience they usually have jobs. Secondly he wasn't laughing - his point was that the best way to fix the issue was to patch
Re:Great (Score:3, Insightful)
Stop a mo and take a deep breath.
Take a look at your (heavily patched I'm sure) machine. If you had an unsupervised 24 hours with that box and an unprivileged account how many other actually useful exploits could you run for key retrieval.
I can think of about 5 methods right now that are much more likely to yield results within Linux. Even on a very up-to-date system with a decent admin.
Pfft.
- Storm in a teacup
Re:Great (Score:5, Informative)
Did you read a paper other than the one I read? I ran the exploit once, taking under one second, and I retrieved enough information to factor the RSA modulus N.
wow, a paper, no kidding (Score:3, Funny)
Well, it's obvious that he has to be right then, since he has published a paper on the topic, right ? Right ? Nobody else can "understand a problem", only him, since he's got a paper on it. A real paper.
All the cryptographers in the world... (Score:2, Funny)
Who would understand what they are saying anyway?
The man (Score:3, Insightful)
I'm sure that if an exploit is found, someone will have a patch ready for the next kernel revision - that's the beauty and advantage of Linux.
Perhaps crypto is not your biggest problem (Score:2)
If security matters, don't do crypto in Linux (Score:5, Insightful)
... or in any other general-purpose operating system on a general-purpose computer. PCs are fundamentally insecure. There are a dozen ways to spy on cryptographic operations done in them, ranging from trojans, to hardware side-channel attacks, and dozens more to get copies of keys that they store. This is just one particular attack that may permit an attacker who can't get a trojan running with sufficient privileges to spy on operations directly to obtain some key bits. But if the attacker can't do that, there are lots of other ways to get the keys. General-purpose computers are simply not trustworthy.
If security is important, you do your crypto in a secure crypto module, like the FIPS 140-2 Level 4 IBM 4758 [ibm.com] or the Level 3 Luna SA [safenet-inc.com]. Or, you use a general-purpose computer with special-purpose, very simple software and then provide strict physical access control to the machine and very limited network access -- often through a serial link using a custom protocol rather than via a real network. Or you could theoretically use a general-purpose machine with a TCPA chip with a regular, general-purpose operating system that has been modified to make use of the TCPA chip and with keys tightly bound to a well-defined system software configuration. But only if you have good physical security. In many situations it's still better to use a FIPS 140-2 Level 3 or Level 4 device.
IMO, the existence of weaknesses like this in Linux, and the fact that they're widely known, is a *good* thing, because it helps convince people not to trust that which is inherently untrustworthy. We need more publicity of similar problems in Windows (and there are lots of them).
Re:If security matters, don't do crypto in Linux (Score:3, Insightful)
Re:If security matters, don't do crypto in Linux (Score:4, Insightful)
Specialist crypto hardware is too expensive for most users
If your secrets are worth enough that an attacker would go to the effort required to use this exploit... no, it is not too expensive. An IBM 4758 costs just over $1000 and provides an extremely high level of security.
If you want to go really cheap, and don't need high performance, you can also use a $10 smart card with a $10 smart card reader. The result is less secure than with a 4758, or a Luna SA, or any of the other products on the market, in several ways, but it's much, much better than doing the crypto in main memory on a GP CPU.
and keeping a PC in a physically secure environment can be difficult and contradictory with other important considerations.
Again, if you need security you have to do what it takes to get security. If security requirements contradict other requirements, you have a problem to solve.
At bottom, the point is that this is a sophisticated attack. There are much easier ways to dig keys out of a GP computer. If you need to defend against this attack, then you also need to defend against all of those, and defending against all of those requires physical security and use of secure crypto hardware. By the time you've closed all the other avenues of attack, you've closed this one, too.
This problem is real, and it's a good idea to fix it, on principle. But it's not uncommon that real security holes are simply irrelevant because of other considerations.
This is analogous to someone pointing out that my house is insecure because the attic vents are made of lightweight sheet metal. An attacker can use a ladder to climb up, use a crowbar to rip the vent off, climb into my attic, find the access panel and drop down into my house. That's all very true. However, compared to all of the *other* ways there are to break into my house, it's simply irrelevant. My house is simply not an appropriate place to store the Hope diamond, and replacing the attic vents with heavy steel ones, or otherwise securing access to the attic, isn't going to change that.
Fix the applications (Score:5, Informative)
This appears to be an application bug, not a kernel one. The kernel never claims to completely isolate processes from one another; though there are memory protections, there are loads of ways that processes can observer each other's actions. This is just a particularly high-resolution one.
The real "bug" here, IMO, is that openSSL believes that no other process can observe anything about its secret computations. Timing attacks against RSA have been known for some time, particularly with regard to modular exponentiation.
It wouldn't be too hard to make RSA encryption take the same amount of time no matter what code path is used, and to make its memory access patterns uncorrelated with the keys (perhaps by using randomization during allocation). They should do this--the fact that their application leaks information has nothing to do with the processor it's running on; it's just that HT makes it particularly easy to measure that information. This would have a performance penalty, and I think the OpenBSD folks are too obsessed with performance, and that's why they've not done this. The performance obsession is a serious problem in the Unix world, and software systems in general.
If implementing openSSL effectively means adding special kernel support for things like constant-length timeslices or cache invalidation between context switches, that's fine. But this is not a bug in the kenel unless the kernel purports to enforce total separation between processes, which it certainly does not.
Re:Fix the applications (Score:3, Interesting)
I don't doubt it, and it would likely take a rewrite of the RSA code. But RSA is a very simple algorithm, so I believe it can be done without much trouble if one doesn't obsess over efficiency.
Practically speaking, just fixing modular exponentiation (which is the core of encoding and decoding) to call exactly the same bn_ functions whether the key bit is 0 or 1 would seem to fix this particular problem.
Colin Percival only cryptographer in whole world (Score:3, Insightful)
It wouldn't be hard to have an option to prevent processes with different owners from running on the same physical CPU at the same time. It wouldn't even affect the case that Linus mentioned. But cryptographers don't seem to think it's a plausible attack anyway, aside from carefully arranged conditions. The discussion was entirely over whether it would be less foolish to prevent it in the kernel or in userspace, and nobody seems to have argued that anything should be done at all.
How to fix this (Score:4, Interesting)
First, remember that public key exchange is usually a very small component of a cryptographic session. Usually, you do all the public key work at startup, exchange private keys for the session, and then just use the private keys. So an inefficient solution that protects the public key computations is acceptable.
In the paper, Percival writes "Further, OpenSSL utilizes a "sliding window" method of modular exponentiation, decomposing x := (a^d)
mod p into a series
of squarings x := x^2 mod p and multiplications x := x ^ (a*2*k+1) mod p." When there's a zero bit in d, the squaring and multiplication are unnecessary, and the SSL implementation apparently skips them. That creates the covert channel. It should be sufficient to revise the code so that it always does the squaring and the multiplication, and then uses the zero bit in D to decide whether or not to use the result. Because some superscalar CPUs might manage to look ahead and avoid the squaring and computation, a nice solution would be to compute both x := (a^d)
mod p and x := (a^(d XOR K) mod p, where K is all binary ones out to the maximum length of d. This makes the computation symmetrical, independent of the value of d. This should roughly double the cost of the public key computation, which is probably tolerable.
It's now important to also examine private key mechanisms for similar vulnerabilities. In particular, DES implementations should be checked. Because those execute millions of times during a long data transfer, even if there's a low-bandwidth leak, an exploit may be possible. Some lookup-table based implementations of DES might have exploitable cache semantics, and that needs to be explored.
Re:Dictator? (Score:5, Funny)
Re:probably easy to fix (Score:4, Informative)
Two reasons: First, I was busy informing vendors -- Intel, Microsoft, the *BSDs, and Linux vendors -- about the problem and explaining the possible solutions. Second, because fixing every application in the world (this doesn't only affect cryptography) would take far more than 3 months.
Re:probably easy to fix (Score:3, Insightful)