I've been doing some research on the Infineon vulnerability known as ROCA over the last few days. As I understand it, the vulnerability is present when a TPM running vulnerable firmware generates an RSA key. At that point, the public key can be used to derive the private key. My question, however, is how this affects secure boot in current versions of Windows. As far as I know, the TPM does not perform any key generation for secure boot and the secure boot keys are managed by Microsoft. Therefor, as far as the end user is concerned, ROCA isn't really relevant in the context of secure boot. Is this correct?

  • I believe you're correct in saying that ROCA is NOT relevant in the context of secure boot, hopefully someone can 100% confirm. – J.J May 14 at 2:43
  • Each TPM has a built-in key which is necessary for integrity. If that key is created with the flawed generation method, then it would be relevant. – forest May 14 at 4:07

As far as I know, the vulnerable library is used by the Federal Information Processing Standard and by CC EAL5+, the same library is used in Windows 10 Secure Boot.

The RSA public key is made of two numbers N (2024 bits) and E (either 317, or commonly E = 2^16 + 1 = 65537), for the sake of performance some implementations use the small E, this implies that E=317 is vulnerable in all key sizes including 2,048-bit keys used in Secure Boot.

Check this advisory published by Microsoft (ADV170012) it lists all Microsoft's affected products.

This to say, YES in some cases ROCA could be relevant in the context of secure boot.

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Not the answer you're looking for? Browse other questions tagged or ask your own question.