[TrouSerS-users] FW: TPM_Seal

[张智 <zha...@ho...>]

From: zha...@ho...
To: haw...@gm...
Subject: RE: [TrouSerS-users] TPM_Seal
Date: Tue, 24 Sep 2013 09:09:24 +0000




 I still do not understand what you've said quite well. The TPM secret that you refer to might be a shared-secret, I guess, which in this case is the owner authorization data stored securely in the TPM when we take ownership.  According to the TPM spec- command part, it is not mentioned that TPM_Seal will use the secret value encrypted in that sealed blob.

Besides, according to the TPM programming guide, there is a maintenance procedure that can be done with the cooperation of both the owner of the TPM and the manufacturer of the TPM, which can restore the SRK to another system.Can we make use of this maintenance method to restore SRK to another system and then unseal the sealed blob ?   

henry

Date: Mon, 23 Sep 2013 23:06:50 -0600
Subject: RE: [TrouSerS-users] TPM_Seal
From: haw...@gm...
To: zha...@ho...

The TPM secret is a value created when you take ownership. It's stored in the protected area of the TPM and never revealed. It is separate from PCRs. Also, it is erased when ownership is cleared. This secret is never shared with any other TPM, or even saved to work between multiple ownerships of the same TPM.

On Sep 23, 2013 10:58 PM, "张智" <zha...@ho...> wrote:




Thanks. 
TPM's secret?  Do you mean the values of the corresponding PCRs ? If it is so, owner can set the same value on another TPM of his, which means that another TPM can have the same secret value as that provided by previous TPM. 


Date: Mon, 23 Sep 2013 22:35:54 -0600
Subject: Re: [TrouSerS-users] TPM_Seal
From: haw...@gm...
To: zha...@ho...

CC: tro...@li...; kj...@us...

No. Sealed data can only be unsealed by the TPM that sealed it. Although storage keys, which are used to seal, can be migratable, sealed data has a secret value encrypted in that is only known to the TPM that performed the seal operation. Per spec, if a TPM successfully decrypts a sealed blob (because it has the private storage key to do so), if the secret inside doesn't match that TPM's secret, it should not release the decrypted data.





Hi, everyone,
      I apologize for this seeminly beginner question, but I can not figure it out after reading TPM spec. 
      The thing is:  when an encryption key which encrypts a file is sealed by executing TPM_Seal on a TPM chip, is it possible to unseal the sealed blob on another TPM chip ? considering that the indicies provided and the values of the corresponding PCRs at the time are the same with that when Seal is invoked.


      We may argue that the key that TPM_Seal uses is non-migratable and thus the sealed blob migration is not available, what if the non-migratable key is wrapped with migratable keys, then it would become migratable ?


      If someone would like to offer me some help, I really appreciate it.
      Thanks a lot !

Best regards,
henry
 		 	   		  

------------------------------------------------------------------------------

October Webinars: Code for Performance

Free Intel webinars can help you accelerate application performance.

Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from

the latest Intel processors and coprocessors. See abstracts and register >

http://pubads.g.doubleclick.net/gampad/clk?id=60133471&iu=/4140/ostg.clktrk
_______________________________________________



TrouSerS-users mailing list

Tro...@li...

https://lists.sourceforge.net/lists/listinfo/trousers-users