On Tue May 21, 2024 at 8:47 AM EEST, Bharat Bhushan wrote:
>
>
> > -----Original Message-----
> > From: Jarkko Sakkinen <jar...@kernel.org>
> > Sent: Tuesday, May 21, 2024 8:47 AM
> > To: Herbert Xu <herb...@gondor.apana.org.au>
> > Cc: linux-integr...@vger.kernel.org; keyri...@vger.kernel.org;
> > andreas.fu...@infineon.com; James Prestwood <prest...@gmail.com>;
> > David Woodhouse <dw...@infradead.org>; Eric Biggers
> > <ebigg...@kernel.org>; James Bottomley
> > <james.bottom...@hansenpartnership.com>; Jarkko Sakkinen
> > <jar...@kernel.org>; David S. Miller <da...@davemloft.net>; open
> > list:CRYPTO API <linux-crypto@vger.kernel.org>; open list <linux-
> > ker...@vger.kernel.org>; Peter Huewe <peterhu...@gmx.de>; Jason
> > Gunthorpe <j...@ziepe.ca>; James Bottomley
> > <james.bottom...@hansenpartnership.com>; Mimi Zohar
> > <zo...@linux.ibm.com>; David Howells <dhowe...@redhat.com>; Paul Moore
> > <p...@paul-moore.com>; James Morris <jmor...@namei.org>; Serge E. Hallyn
> > <se...@hallyn.com>; open list:SECURITY SUBSYSTEM <linux-security-
> > mod...@vger.kernel.org>
> > Subject: [EXTERNAL] [PATCH v2 5/6] tpm: tpm2_key: Extend parser to
> > TPM_LoadableKey
> >
> > ----------------------------------------------------------------------
> > Extend parser to TPM_LoadableKey. Add field for oid to struct tpm2_key
> > so that callers can differentiate different key types.
> >
> > Signed-off-by: Jarkko Sakkinen <jar...@kernel.org>
> > ---
> > drivers/char/tpm/tpm2_key.c | 14 +++++++++++---
> > include/crypto/tpm2_key.h | 2 ++
> > security/keys/trusted-keys/trusted_tpm2.c | 4 ++++
> > 3 files changed, 17 insertions(+), 3 deletions(-)
> >
> > diff --git a/drivers/char/tpm/tpm2_key.c b/drivers/char/tpm/tpm2_key.c
> > index 0112362e432e..59797dc232f1 100644
> > --- a/drivers/char/tpm/tpm2_key.c
> > +++ b/drivers/char/tpm/tpm2_key.c
> > @@ -32,16 +32,24 @@ int tpm2_key_type(void *context, size_t hdrlen,
> > const void *value, size_t vlen)
> > {
> > enum OID oid = look_up_OID(value, vlen);
> > -
> > - if (oid != OID_TPMSealedData) {
> > + struct tpm2_key *key = context;
> > +
> > + switch (oid) {
> > + case OID_TPMSealedData:
> > + pr_info("TPMSealedData\n");
> > + break;
> > + case OID_TPMLoadableKey:
> > + pr_info("TPMLodableKey\n");
These should be pr_debug() (forgot to change).
> > + break;
> > + default:
> > char buffer[50];
> > -
> > sprint_oid(value, vlen, buffer, sizeof(buffer));
> > pr_debug("OID is \"%s\" which is not TPMSealedData\n",
> > buffer);
>
> Maybe extend this print to say "neither TPMSealedData nor TPMLodableKey"
Right, I tried to apply minimal delta to patches where existing code
needs to be carved to a new form :-)
I think it could be just "OID \"%s\" is unknown"?
BR, Jarkko
