On Thu, Dec 06, 2018 at 07:00:13PM +0100, Roberto Sassu wrote: > On 12/5/2018 12:53 AM, Jarkko Sakkinen wrote: > > On Tue, Dec 04, 2018 at 09:21:36AM +0100, Roberto Sassu wrote: > > > + for (i = 0; i < ARRAY_SIZE(tpm2_hash_map); i++) { > > > + enum hash_algo crypto_algo = tpm2_hash_map[i].crypto_id; > > > + > > > + if (bank->alg_id != tpm2_hash_map[i].tpm_id) > > > + continue; > > > + > > > + bank->digest_size = hash_digest_size[crypto_algo]; > > > + bank->crypto_id = crypto_algo; > > > + return 0; > > > + } > > > + > > > + return tpm2_pcr_read(chip, 0, &digest, &bank->digest_size); > > > +} > > > > This is a part that I don't get. Coud you just always call > > tpm2_pcr_read() instead of this complexity > > First, we avoid operations that may increase the boot time. Second, the > loop is necessary to obtain the crypto subsystem identifier from a TPM > algorithm identifier. I think here would be a comment in place and it would be fine. /Jarkko