Re: [PATCH v2 1/8] tss: Fix handling of TPM_RH_NULL in intel-tss

["Jarkko Sakkinen" <jarkko@kernel.org>]

On Sun Aug 4, 2024 at 4:42 PM EEST, James Bottomley wrote:
> The design of the intel-tss shim is to hide the difference between the
> internal and the external handles by doing the internal to external
> transform on entry.  Unfortunately, the NULL handle (TPM_RH_NULL,
> 40000007) has two possible internal representations depending on
> whether it's used to indicate no session or the null hierarcy.
>
> There is a bug in the intel-tss in that it uses the wrong internal
> NULL handle to try to create the NULL seed primary (and thus fails).
> Now that we're going to be using the NULL primary to salt sessions,
> the Intel TSS shim needs fixing to cope with thi correctly.
>
> The fix is to do the correct transform to the internal hierarchy
> representation on NULL hierarchy creation and to do the session handle
> conversion everywhere else.  Additionally remove the intel_handle()
> code which was supposed to do this: it's unused because 0 is never
> passed in as a handle number.
>
> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
>
> ---
> v2: reword commit message
>
> ---
>  src/include/intel-tss.h | 18 +++++-------------
>  1 file changed, 5 insertions(+), 13 deletions(-)
>
> diff --git a/src/include/intel-tss.h b/src/include/intel-tss.h
> index 1870b4e..5b8db20 100644
> --- a/src/include/intel-tss.h
> +++ b/src/include/intel-tss.h
> @@ -251,14 +251,6 @@ intel_sess_helper(TSS_CONTEXT *tssContext,
> TPM_HANDLE auth, TPMA_SESSION flags)
>  				  TPMA_SESSION_CONTINUESESSION |
> flags);
>  }
>  
> -static inline TPM_HANDLE
> -intel_handle(TPM_HANDLE h)
> -{
> -	if (h == 0)
> -		return ESYS_TR_NONE;
> -	return h;
> -}
> -
>  static inline void
>  TSS_Delete(TSS_CONTEXT *tssContext)
>  {
> @@ -937,8 +929,10 @@ tpm2_CreatePrimary(TSS_CONTEXT *tssContext,
> TPM_HANDLE primaryHandle,
>  	TPM2B_PUBLIC *opub;
>  	TPM_RC rc;
>  
> -	/* FIXME will generate wrong value for NULL hierarchy */
> -	primaryHandle = intel_handle(primaryHandle);
> +
> +	/* TPM_RH_NULL is mapped to ESYS_TR_NONE, which won't work
> here */
> +	if (primaryHandle == TPM_RH_NULL)
> +		primaryHandle = INT_TPM_RH_NULL;
>  
>  	outsideInfo.size = 0;
>  	creationPcr.count = 0;
> @@ -993,9 +987,7 @@ tpm2_StartAuthSession(TSS_CONTEXT *tssContext,
> TPM_HANDLE tpmKey,
>  		      TPM_HANDLE *sessionHandle,
>  		      const char *bindPassword)
>  {
> -	bind = intel_handle(bind);
> -	tpmKey = intel_handle(tpmKey);
> -	if (bind != ESYS_TR_NONE)
> +	if (bind != TPM_RH_NULL)
>  		intel_auth_helper(tssContext, bind, bindPassword);

Not blaming the patch but just have hard time coping this.

The most basic question is probably this: what is the application for
INT_TPM_RH_NULL?

Let's imagine that we have a flow chart describing Intel shim as a state
machine. I decide to shoot it with these three stimulus:

1. INT_TPM_RH_NULL
2. TPM_RH_NULL
3. A valid handle

What happens in each case to its state?

>  
>  	return Esys_StartAuthSession(tssContext, tpmKey, bind,
> ESYS_TR_NONE,

BR, Jarkko