Re: [tpmdd-devel] [PATCH RFC v4 0/5] RFC: in-kernel resource manager

[Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>]

On Mon, Jan 23, 2017 at 09:01:03AM -0800, James Bottomley wrote:
> On Mon, 2017-01-23 at 01:44 +0200, Jarkko Sakkinen wrote:
> > This patch set adds support for TPM spaces that provide a context
> > for isolating and swapping transient objects. The content does
> > not yet include support for policy and HMAC sessions.
> > 
> > There's a test script for trying out TPM spaces in
> > 
> >   git://git.infradead.org/users/jjs/tpm2-scripts.git
> > 
> > A simple smoke test can be run by
> > 
> >   sudo python -m unittest -v tpm2_smoke.SpaceTest   
> > 
> > v2:
> > Changed to James' proposal of API. I did not make any other changes
> > except split core TPM space code its own patch because I want to find
> > consensus on the API before polishing the corners. Thus, this version
> > also carries the RFC tag. I have not yet locked in my standpoint
> > whether
> > ioctl or a device file is a better deal.
> > 
> > v3:
> > - Check TPM return code in tpm_map_response.
> > - Reference tracking for /dev/tpms0.
> > - clear_bit(is_open) was removed from tpm-dev.c. Added it back.
> > - Use response length as the buffer size limit in tpm2_commit_space.
> > - This version now passes again my smoke tests.
> > 
> > v4:
> > - Lots of small bug fixes and clean ups.
> > - Quirk for TPM2_CC_FlushHandle
> 
> It's still failing my flush test.  This time the problem is the return
> code on context save failure: it's TPM_RC_REFERENCE_H0 not
> TPM_RC_HANDLE.  This is the fix.  The manual implies TPM_RC_HANDLE
> could also be the return, so I kept both.
> 
> James
> 
> ---
> 
> diff --git a/drivers/char/tpm/tpm.h b/drivers/char/tpm/tpm.h
> index 62e8421..cc1db77 100644
> --- a/drivers/char/tpm/tpm.h
> +++ b/drivers/char/tpm/tpm.h
> @@ -106,6 +106,7 @@ enum tpm2_algorithms {
>  	TPM2_ALG_SHA512		= 0x000D,
>  	TPM2_ALG_NULL		= 0x0010,
>  	TPM2_ALG_SM3_256	= 0x0012,
> +	TPM2_RC_REFERENCE_H0	= 0x0910,
>  };
>  
>  enum tpm2_command_codes {
> diff --git a/drivers/char/tpm/tpm2-space.c b/drivers/char/tpm/tpm2-space.c
> index 83e9708..b36337a 100644
> --- a/drivers/char/tpm/tpm2-space.c
> +++ b/drivers/char/tpm/tpm2-space.c
> @@ -104,7 +104,8 @@ static int tpm2_save_context(struct tpm_chip *chip, u32 handle, u8 *buf,
>  			 __func__, rc);
>  		tpm_buf_destroy(&tbuf);
>  		return -EFAULT;
> -	} else if ((rc & TPM2_RC_HANDLE) == TPM2_RC_HANDLE) {
> +	} else if ((rc & TPM2_RC_HANDLE) == TPM2_RC_HANDLE ||
> +		   rc == TPM2_RC_REFERENCE_H0) {
>  		tpm_buf_destroy(&tbuf);
>  		return -ENOENT;
>  	} else if (rc) {

I applied and pushed a fix that checks just TPM_RC_REFERENCE_H0.

/Jarkko