Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

The GPU, not the TPM, is the root of hardware DRM

mjg59.dreamwidth.org

Post reply on HN
Print view View original

231–240 of 493 posts

Re: The GPU, not the TPM, is the root of hardware DRM

#231

Post by cryptonector »

mjg59 wrote: post #139
cryptonector wrote: post #137

> Now, TPMs are sometimes referred to as a TEE, and in a way they are. However, they're fixed function - you can't run arbitrary code on the TPM, you only have whatever functionality it provides. But TPMs do have the ability to decrypt data using keys that are tied to the TPM, so isn't this sufficient? Well, no. First, the TPM can't communicate with the GPU. The OS could push encrypted material to it, and it would ge…

How would the GPU verify it's speaking to a real TPM? You'd need to bake the full set of legitimate EK cert CAs into it somehow (charitably let's say that's a signed blob that the driver pushes in at startup), but that's still going to be a terrible user experience because you won't get media playback if your machine has a TPM that's too new or from too niche a vendor.

> How would the GPU verify it's speaking to a real TPM?

Option 1: as I said, the GPU could have its own, and yes in that case the EK cert would be known to the GPU (or it could have a platform-like cert issued by the GPU OEM).

Option 2: the platform vendor can teach the GPU the EK cert (or the public key for some primary key anyways).

Option 3: the GPU could learn it on first use.

> charitably let's say that's a signed blob that the driver pushes in at startup

That's what TPM vendors do as to the EK cert. Surely if they can do that then so can GPU and platform vendors. Indeed, some platform vendors ship with platform certs.

> but that's still going to be a terrible user experience because you won't get media playback if your machine has a TPM that's too new or .

What do you mean "too new"? Like, you replaced your TPM? That's a thing on servers, but not laptops.

As to "from too niche a vendor", as long as the platform vendor teaches the GPU what the EK cert is, or makes a platform-like cert that the GPU can use to authenticate the TPM, then it's good enough.

Anyways I suspect that MSFT and others don't mind an incrementalist approach. You have a system that can do it their way? Great, it will. You have a system that cannot do it their way? Fine, they'll do weak software DRM for now. There's probably no other way to to get to their dream DRM everywhere state.

Re: The GPU, not the TPM, is the root of hardware DRM

#232

Post by alex7734 »

andrewaylett wrote: post #224
2OEH8eoCRo0 wrote: post #218

Earlier quoted context omitted.

I thought it was to DRM media?

As the article points out, the TPM is not in a good place, architecturally, to use for DRM: there's no path from the TPM to the screen that's not under OS (and thereby user) control.

The whole point of TPM is that the OS is not under the user control anymore.

If you modify it thanks to remote attestation you can no longer prove that it is unmodified using the TPM.

Re: The GPU, not the TPM, is the root of hardware DRM

#233

Post by cryptonector »

mike_hearn wrote: post #176
mjg59 wrote: post #139

Earlier quoted context omitted.

How would the GPU verify it's speaking to a real TPM? You'd need to bake the full set of legitimate EK cert CAs into it somehow (charitably let's say that's a signed blob that the driver pushes in at startup), but that's still going to be a terrible user experience because you won't get media playback if your machine has a TPM that's too new or from too niche a vendor.

Right, and more to the point there's nothing special about a TPM design-wise. It's actually a very odd kind of chip that only really exists due to the unique political and market requirements of the PC industry. If you look at vertically integrated platforms like Apple's, or the games consoles, or smartphones, there's no TPM. There are subsystems that do similar things, but none of them follow the TPM design specs. E…

TPM is a missed opportunity. What I really want for security is a solid secure enclave scheme on the CPU itself so my SE code can blaze. The TPM is not programmable and is very limited, both in terms of its API and in terms of its capabilities (e.g., number of keys loaded, number of algorithms supported, ...) and in terms of its performance.

My point in my above reply was to say that even if TPMs were used by GPUs then TFA's point would still stand.

Re: The GPU, not the TPM, is the root of hardware DRM

#234

Post by talldayo »

stackghost wrote: post #60
talldayo wrote: post #24

Earlier quoted context omitted.

I think that's a misunderstanding of what the FSF stands for overall, though. The FSF can never be a diplomatic negotiator for the benefit of free software; they are idealists, even when it serves against their own interests. Their whole shtick is not settling for half-baked appeasements, and so they're destined to be a pariah of the tech industry at-large. Neither you nor me can stop them, it's entirely within their…

There's no misunderstanding on my part; it's why I said that their ignorance is totally on-brand. >more like the mainstream has abandoned free software. Indeed, because free software development is largely driven by ideological purity rather than feature parity. Mainstream users see Free Software people as irrelevant kooks, and thus easy to dismiss, which is why Free Software has so utterly failed as a movement. >You…

> Indeed, because free software development is largely driven by ideological purity rather than feature parity.

Ideological purity is a valuable thing. Look at Minix, hell, even look at the BSDs today. These are projects that have collapsed because of their feature obsession and ignorance of ideology. The differentiation of ideology is what makes free software uniquely successful - it is the feature.

> Mainstream users see Free Software people as irrelevant kooks, and thus easy to dismiss, which is why Free Software has so utterly failed as a movement.

Mainstream users don't think about Free Software at all. They certainly use it though. They rely on it, to provide and maintain the runtime their cell phone and iPad and router all depend on. It probably runs an RTOS on their grandpa's CPAP machine, it probably occupies the DVR for their cable TV and it's likely running on their games console and personal computer, too.

Free software is even more inescapable than proprietary software. If users cared enough to understand the difference, you and I both know they would accuse the businesses of being the irrelevant kooks. Not a single "maintream user" I know would defend Apple or Google or Microsoft's business practices as software companies. No one.

> I'm sure flat-earthers feel that my belief that earth is an oblate spheroid is peculiar, too. Of what relevance is that to anyone?

As the other comment suggested, this is both an insincere response and one where you are the flat earther here. The FSF has reasons that they hold the principles they do, and you haven't refuted any of their ideology. You are the guy lambasting Gallileo, and when Gallileo asks you why heliocentrism offends you, you are replying "because the mainstream clergy sees you as kooks." It's not a response at all.

> The emacs developers don't even understand how large chunks of emacs work

Nobody is so stupid that we expect every kernel dev to understand the whole of the kernel. It's folly, and not what I was asking anyways. Nobody at Apple understands how the entirety of iOS works either, but that's not an implication that it's inherently insecure. What makes the FSF balk at Apple is the inaccountability. The lack of reason associated with their statements asserting the privacy and security of a system that sues it's auditors.

If you have a more reasonable heuristic to suggest, I'm all ears.

Re: The GPU, not the TPM, is the root of hardware DRM

#235

Post by ta1243 »

happymellon wrote: post #210
sidewndr46 wrote: post #204

Earlier quoted context omitted.

You're at an industry conference. I want the data on your laptop's hard drive. You leave your laptop in the hotel room. Which one is easier: 1. Go into your room and screw around with the boot loader to somehow give me unencrypted access to your laptop after you login next time. 2. Go into your room. Take your laptop. Put an identical looking laptop in place that runs software that boots and looks identical. Have it…

Passwords are generally defeated by a hammer to the fingers. Repeat until password is extracted.

Knees probably better -- break my fingers and I can't give you my password

Re: The GPU, not the TPM, is the root of hardware DRM

#236

Post by p_ing »

m463 wrote: post #149
osy wrote: post #26

> I'm going to be honest here and say that I don't know what Microsoft's actual motivation for requiring a TPM in Windows 11 is. It is quite obvious: to force people to buy a new PC. TPM provides no added security value for the vast majority of users[1] but it is a convenient hardware that has only started to become standard (fTPM) in PCs built in the last ~8 years so it provides an excuse for Microsoft to declare co…

Microsoft is just trying to match features with apple which does the sorts of things with the T2 chip. Home users probably don't care that much, but corporate users do. That said, the root of all DRM is not the TPM or the GPU or whatever... it is hollywood.

You mean Apple matched features with Microsoft.

Devices with dTPM were released in 2006. BitLocker leveraging dTPM released with Windows Vista. Corporations have been using BitLocker w/ TPM for nearly two decades at this point.

Re: The GPU, not the TPM, is the root of hardware DRM

#237

Post by 2OEH8eoCRo0 »

alex7734 wrote: post #232
andrewaylett wrote: post #224

Earlier quoted context omitted.

As the article points out, the TPM is not in a good place, architecturally, to use for DRM: there's no path from the TPM to the screen that's not under OS (and thereby user) control.

The whole point of TPM is that the OS is not under the user control anymore. If you modify it thanks to remote attestation you can no longer prove that it is unmodified using the TPM.

Do they mean that no OS modification is necessary to read the decrypted media from memory?

Re: The GPU, not the TPM, is the root of hardware DRM

#238

Post by alerighi »

AnthonyMouse wrote: post #127
p_ing wrote: post #103

Earlier quoted context omitted.

> TPM provides no added security value for the vast majority of users[1] Yes it does. The vast majority of users aren't going to have their laptop stolen by the CIA/NSA and have their DIMMs popped and cryofreezed. The vast majority of users aren't going to have the case opened and a special-purpose PCIe device installed to steal keys over DMA. The vast majority of users aren't going to have a dTPM vulnerable to SPI s…

> The vast majority of users aren't going to have their laptop stolen by the CIA/NSA and have their DIMMs popped and cryofreezed. That's kind of the point. The vast majority of users aren't going to have their laptop stolen at all, if they do it will 99% of the time be by someone who only wants to wipe it and fence it, and attempts to access data are most likely to be by unsophisticated family members who would be de…

The vast majority of users neither have a password on their computer, or if they have it it's a stupid one (like their name, their birthday, etc) or they have it written on a post-it that is attached on the monitor itself. Why do they need a TPM? Most of the time I setup a computer for a friend or family member they ask me to remove the password since they don't want to remember it.

Vast majority of users neither have that much important data to steal on their computer at all, just some family photos, some movies downloaded from the internet, there is the case of credentials saved in the browser, but the most important stuff (such as banking sites) nowadays requires a multiple factor authentication (such as password + OTP on your phone) to do any operation.

Re: The GPU, not the TPM, is the root of hardware DRM

#239

Post by __m »

saxonww wrote: post #87
zb3 wrote: post #19

Earlier quoted context omitted.

Integrated GPUs exist. Wouldn't it make more sense that the "high value" content should not be exposed to any external GPU? Then we can treat those integrated ones as part of the "TEE". That's my speculation, waiting for details.

This is the question I had about this. The reason this design works per the article is that the GPU memory is inaccessible to the OS, so the decrypted content cannot be stolen. With a unified memory architecture, is the shared GPU memory inaccessible to the CPU?

The GPU ultimately has to output unencrypted content, it will always be possible to steal unless we manage to implement drm in human eyes

Re: The GPU, not the TPM, is the root of hardware DRM

#240

Post by pxoe »

AnthonyMouse wrote: post #122
jsheard wrote: post #115

Earlier quoted context omitted.

DD2 is a single player game, those generally don't maintain their active player counts forever. It peaked at 228,285 concurrent (not total) Steam players which are pretty good numbers.

The peak was the release, after which it promptly cratered. It was below half that within a month and below 14k the next month.

This obsession with concurrent player counts, especially for single player games, is just obtuse. It's not actually telling what you want to believe.
Post reply on HN