HDCP & Paranoid Copy Protection Wires

I’ve come to really enjoy the ROKU device and service. Now mostly we just watch Netflix on it, and the LG has a Netflix app of it’s own, so for that TV it is kind of redundant. However, even there, it has a couple of benefits. First off, it is faster. The built in LG App takes enough resources that whatever passes as a brain in the “Smart” part of the LG “smart” TV can’t walk and chew gum at the same time. So, for example, if you re running the Netflix LG App, some of the buttons on the remote can no longer work. I’ve not catalogued exactly which, but things like changing the picture attributes and such ( i.e. changing volume works, but don’t try to get into or out of ‘sports mode’ video enhancement IIRC). So most of the time we use the Roku stick on the LG and ignore its “smarts”.

The ROKU is also a distinct gadget, so can travel with me to hotel rooms (provided they have fast WiFi…) and it can simply be unplugged from the TV to assure it isn’t doing anything. The Toshiba “dumb” TV with a Roku stick is preferable to the LG with Apps built in. (The sound on the Toshiba is also better. At higher volumes, the LG starts to have some “buzz” from the plastics vibrating. The Toshiba seems to have better Dolby decode as well.) So from a security point of view, dumb + stick beats “built in and can’t be sure it is off”. Were I having a ‘do over’ I’d get two Toshibas instead.

Now on the Roku there are so many channels it is a few weeks work to look at them all and decide, so I’ve not done that. That isn’t including the private channels that need a special password to see, either. (Yes, you can set up your own family TV station only shared with those you give the magic words to…) I’ve got a very very small subset selected ( 333 channels…) and I’ve at least done a couple of minute sample of each of them to verify they might have some interest. (i.e. don’t demand a ‘registration’ or want me to pay to view, and seem to have a theme that might interest me). Realize, too, that each of those “channels” can have subdirectories or subchannels, so things like Pluto have 100ish other stations, like RT and Newsy and NBC and…, lurking to explore further. I’ve still not looked at most of the 333 for longer than my initial selection minute.

Along the way, the ones I have looked at, are often what seems to be someone’s old VHS library uploaded. Many channels have the same selections of movies where the copyright has expired. So I need to prune that list down to a few that I like, too. Lots of old John Wayne Westerns repeat, for example. But while watching several of these “cheap and sleazy” providers, I’d noticed an odd “flash” of an image on the screen. WT?

Usually only one brief time, and a line of text I couldn’t read as I usually was looking elsewhere in the frame when it popped up. Eventually one channel was “bad enough” it popped about every 4 minutes. I decided to stare successively at where each word showed up and “get the message”. It’s a white text on a purplish field. It is an HDCP copy violation notice. “HDCP Unauthorized Content”

OK, so I’d thought maybe these folks had boosted a copy of their stuff and left in this to warn they were boosted, or maybe they didn’t realize.

I was wrong.

Looking into that message, I found it was caused when “your HDMI cable is too long”, among other things. But I didn’t have a cable. The Roku plugged directly into the TV…

A bit more digging, and I found that in order to thwart copy making, the HDMI standard includes a protocol for validating that the devices on each end are ‘approved’ and have signed up to not allow copies to be made. IF your “device” doesn’t compute the right magic cookie encryption key fast enough, you get a message saying Your Have Been A Bad Boy!!!… Now since the Roku must talk over the internet to get the ‘credentials’ of the supplier to compute the magic sauce, it’s a little slower than your Satellite Box at coming up with the key. For most sites, this isn’t an issue, but it looks like some of the cheaper ones are running on servers that just don’t answer as fast as a real service (like Netflix and Youtube and The CW) so you get that message flash, sometimes, on some of them. OK, I can live with that, I guess.

https://support.roku.com/article/208755468-i-see-a-purple-screen-that-says-hdcp-unauthorized-content-disabled-when-i-try-to-play-content-

What to do if you see a purple screen or an “HDCP Unauthorized” message
Background

A Roku® streaming player displays a purple screen as shown below with the message “HDCP unauthorized.”, or just a plain purple screen, if it detects that the TV or Audio/Video Receiver (AVR) does not support the proper content protection technology, called HDCP. HDCP is the standard the movie and TV industry requires for preventing copying of digital audio and video content over an HDMI® cable link. If your TV or AVR does not support HDCP and you try to play video, you may see one of the purple screens below. This issue may also be a result of other problems like a faulty HDMI cable or HDMI connector. This FAQ explains how to resolve the issue of seeing a purple screen when using your Roku player.

This explains what to do if it comes and stays, but they don’t seem to care about the ‘flicker’ on some channels where it shows and goes…

And, of course, there’s a wiki…

https://en.wikipedia.org/wiki/High-bandwidth_Digital_Content_Protection

High-bandwidth Digital Content Protection (HDCP) is a form of digital copy protection developed by Intel Corporation[1] to prevent copying of digital audio and video content as it travels across connections. Types of connections include DisplayPort (DP), Digital Visual Interface (DVI), and High-Definition Multimedia Interface (HDMI), as well as less popular, or now defunct, protocols like Gigabit Video Interface (GVIF) and Unified Display Interface (UDI).

The system is meant to stop HDCP-encrypted content from being played on unauthorized devices or devices which have been modified to copy HDCP content. Before sending data, a transmitting device checks that the receiver is authorized to receive it. If so, the transmitter encrypts the data to prevent eavesdropping as it flows to the receiver.

In order to make a device that plays HDCP-enabled work, the manufacturer must obtain a license from Intel subsidiary Digital Content Protection LLC, pay an annual fee, and submit to various conditions. For example, the device cannot be designed to copy; it must “frustrate attempts to defeat the content protection requirements”; it must not transmit high definition protected video to non-HDCP receivers; and DVD-Audio works can be played only at CD-audio quality by non-HDCP digital audio outputs (analog audio outputs have no quality limits).

Oh, gee, security from Intel, the folks who brought your broken math (the “repentium” bug) and insecure coprocessors you can’t control. What could possibly go wrong…

Cryptanalysis researchers demonstrated flaws in HDCP as early as 2010. In September 2010, an HDCP master key that allows for the generation of valid device keys was released to the public, rendering the key revocation feature of HDCP useless. Intel has confirmed that the crack is real, and believes the master key was reverse engineered rather than leaked. In practical terms, the impact of the crack has been described as “the digital equivalent of pointing a video camera at the TV”, and of limited importance for consumers because the encryption of high-definition discs has been attacked directly, with the loss of interactive features like menus. Intel threatened to sue anyone producing an unlicensed device.

Gee, I’m sure suing can provide copy protection… NOT!

But wait, there’s more…

In the United States, the Federal Communications Commission (FCC) approved HDCP as a “Digital Output Protection Technology” on August 4, 2004. The FCC’s Broadcast flag regulations, which were struck down by the United States Court of Appeals for the District of Columbia Circuit, would have required DRM technologies on all digital outputs from HDTV signal demodulators. Congress is still considering legislation that would implement something similar to the Broadcast Flag. The HDCP standard is more restrictive than the FCC’s Digital Output Protection Technology requirement. HDCP bans compliant products from converting HDCP-restricted content to full-resolution analog form, presumably in an attempt to reduce the size of the analog hole.

On January 19, 2005, the European Information, Communications, and Consumer Electronics Technology Industry Associations (EICTA) announced that HDCP is a required component of the European “HD ready” label.

Microsoft Windows Vista and Windows 7 both use HDCP in computer graphics cards and monitors.

Well, yet another reason not to use Microsoft or Windows… Nothing like forcing a broken copy prevention overhead onto everything and then trying to debug it when things don’t work perfectly…

Then there’s the various ways it has been broken, oh, and it comes in “versions” so might it someday suffer an ‘upgrade’ that is NOT backward compatible with your present devices and require you to buy all new ones? Hmmm….

Circumvention

HDCP strippers remove HDCP information from the video signal in order to allow the data to flow freely to a non-HDCP display. It is currently unclear whether such devices would remain working if the HDCP licensing body issued key-revocation lists, which may be installed via new media (e.g. newer Blu-ray Discs) played-back by another device (e.g. a Blu-ray Disc player) connected to it.

Cryptanalysis

In 2001, Scott Crosby of Carnegie Mellon University wrote a paper with Ian Goldberg, Robert Johnson, Dawn Song, and David Wagner called “A Cryptanalysis of the High-bandwidth Digital Content Protection System”, and presented it at ACM-CCS8 DRM Workshop on November 5.

The authors concluded that HDCP’s linear key exchange is a fundamental weakness, and discussed ways to:

Eavesdrop on any data.
Clone any device with only its public key.
Avoid any blacklist on devices.
Create new device key vectors.
In aggregate, usurp the authority completely.

They also said the Blom’s scheme key swap could be broken by a so-called conspiracy attack: obtaining the keys of at least 40 devices and reconstructing the secret symmetrical master matrix that was used to compute them.

Around the same time, Niels Ferguson independently claimed to have broken the HDCP scheme, but he did not publish his research, citing legal concerns arising from the controversial Digital Millennium Copyright Act.

In November 2011 Professor Tim Güneysu of Ruhr-Universität Bochum revealed he had broken the HDCP 1.3 encryption standard.

Master key release

On September 14, 2010, Engadget reported the release of a possible genuine HDCP master key which can create device keys that can authenticate with other HDCP compliant devices without obtaining valid keys from The Digital Content Protection LLC. This master key would neutralize the key revocation feature of HDCP, because new keys can be created when old ones are revoked. Since the master key is known, it follows that an unlicensed HDCP decoding device could simply use the master key to dynamically generate new keys on the fly, making revocation impossible. It was not immediately clear who discovered the key or how they discovered it, though the discovery was announced via a Twitter update which linked to a Pastebin snippet containing the key and instructions on how to use it. Engadget said the attacker may have used the method proposed by Crosby in 2001 to retrieve the master key, although they cited a different researcher. On September 16, Intel confirmed that the code had been cracked. Intel has threatened legal action against anyone producing hardware to circumvent the HDCP, possibly under the Digital Millennium Copyright Act.

HDCP v2.2, v2.1 and v2.0 breach

On August 2012 version 2.1 was proved broken. The attack used the fact that the pairing process sends the Km key obfuscated with an XOR. That makes the encryptor (receiver) unaware of whether it encrypts or decrypts the key. Further, the input parameters for the XOR and the AES above it are fixed from the receiver side, meaning the transmitter can enforce repeating exactly the same operation. Such a setting allows an attacker to monitor the pairing protocol, repeat it with a small change and extract the Km key. The small change is to pick the “random” key to be the encrypted key from the previous flow. Now, the attacker runs the protocol and in its pairing message it gets E(E(Km)). Since E() is based on XOR it undoes itself, thus exposing the Km of the legitimate device.

V2.2 was released to fix that weakness by adding randomness provided by the receiver side. However the transmitter in V2.2 must not support receivers of V2.1 or V2.0 in order to avoid this attack. Hence a new erratum was released to redefine the field called “Type” to prevent backward compatibility with versions below 2.2. The “Type” flag should be requested by the content’s usage rules (i.e. via the DRM or CAS that opened the content).

In August 2015, version 2.2 was rumored to be broken. An episode of Netflix’s UHD version of series Breaking Bad was leaked to the Internet in UHD format; its metadata indicated it was an HDMI cap, meaning it was captured through HDMI interface that removed HDCP 2.2 protection.

On November 4, 2015, Chinese company LegendSky Tech Co., already known for their other HDCP rippers/splitters under the HDFury brand, released the HDFury Integral, a device that can remove HDCP 2.2 from HDCP-enabled UHD works. On December 31, 2015, Warner Bros and Digital Content Protection, LLC (DCP, the owners of HDCP) filed a lawsuit against LegendSky. Nevertheless, the lawsuit was ultimately dropped after LegendSky argued that the device did not “strip” HDCP content protection but rather downgraded it to an older version, a measure which is explicitly permitted in DCP’s licensing manual.

Problems

HDCP can cause problems for users who want to connect multiple screens to a device; for example, a bar with several televisions connected to one satellite receiver or when a user has a closed laptop and uses an external display as the only monitor. HDCP devices can create multiple keys, allowing each screen to operate, but the number varies from device to device; e.g., a Dish or Sky satellite receiver can generate 16 keys. The technology sometimes causes handshaking problems where devices cannot establish a connection, especially with older high-definition displays.

Edward Felten wrote “the main practical effect of HDCP has been to create one more way in which your electronics could fail to work properly with your TV,” and concluded in the aftermath of the master key fiasco that HDCP has been “less a security system than a tool for shaping the consumer electronics market.”

Additional issues arise when interactive media (i.e. video games) suffer from control latency, because it requires additional processing for encoding/decoding. Various everyday usage situation, such as live streaming or capture of game play, are also adversely affected.

There is also the problem that all Apple laptop products, presumably in order to reduce switching time, when confronted with an HDCP compliant device, automatically switch all output from the DVI / Mini DisplayPort / Thunderbolt connector port to HDCP compliant. This is a problem if the user wishes to record or use videoconferencing facilities further down the chain, because these are inherently forbidden by HDCP. This applies even if the output is not HDCP work, like a PowerPoint presentation.

Additionally all Android based devices and some later PC tablets encrypt all content at the output, regardless of whether the user is displaying simple presentation works such as a PowerPoint, or copyrighted works such as a DVD or downloaded movie. This causes issues for recording devices such as those found in all major university lecture rooms. Lecturer’s PowerPoint slides or other non-copyrighted works cannot be captured or displayed in the university lecture recordings.

Well, at least now we know why all those perfectly legal and desirable things to do are no longer working. It is all to “protect” us from having the same ability to make home recordings of shows as we had on our old VHS machines… One wonders if “The Sony Decision” could be used (since it found we had that ‘right’) to show this process is taking away our rights…

Sigh.

Yet Another Copy Protect scheme that doesn’t work and has been entirely compromised, but is still being forced on everyone despite that, and despite the fact that it breaks perfectly legal uses of YOUR stuff.

But at least now I know why those things won’t work, and why I get that “purple flash” sometimes in cheapo channels…

I wonder if there is a Linux based HDCP defeat…

FWIW, the more common approach seems to be a cheap “HDMI Splitter” where some of them “authenticate” on one of the lines but then opens output to both and may (or may not) strip the HDCP from the second port. Cheap, made in China (of course…) and available. For anyone who isn’t interested in trick ways to make their own keys and totally hose the “protection” at the software level.

So much stupid driven by so much greed causing so much broken to no benefit.

Subscribe to feed

About E.M.Smith

A technical managerial sort interested in things from Stonehenge to computer science. My present "hot buttons' are the mythology of Climate Change and ancient metrology; but things change...
This entry was posted in Tech Bits and tagged , , , . Bookmark the permalink.

5 Responses to HDCP & Paranoid Copy Protection Wires

  1. LG says:

    I’ve just learned something new here @ ChiefIO University, where the learning never stops.

  2. Graeme No.3 says:

    Curious that I read this to find a pop-up add for USB and other DVD players.

  3. philjourdan says:

    The MPAA (among others) has abused the DCA to the point where they are no longer in compliance with the laws. But since they wield the influence with congress nothing will be done. Consumers will continue to abandon them until they have nothing left.

    I hate being cheated. Even more so when they laugh at you while they do it.

  4. Steve C says:

    Hey, Damn Clever People!
    Another challenge to test your “skillz” … ;-)

  5. Pingback: Disappointing Odroid C1+ | Musings from the Chiefio

Comments are closed.