| Message ID | 20181030154711.2782-6-roberto.sassu@huawei.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | tpm: retrieve digest size of unknown algorithms from TPM | expand |
On Tue, 30 Oct 2018, Roberto Sassu wrote: > This patch ensures that the digest size returned by the TPM during a PCR > read matches the size of the algorithm passed as argument to > tpm2_pcr_read(). The check is performed after information about the PCR > banks has been retrieved. > > Signed-off-by: Roberto Sassu <roberto.sassu@huawei.com> What is the scenarion when this can happen (should be explained in the commit message)? /Jarkko
On 10/30/2018 8:52 PM, Jarkko Sakkinen wrote: > On Tue, 30 Oct 2018, Roberto Sassu wrote: >> This patch ensures that the digest size returned by the TPM during a PCR >> read matches the size of the algorithm passed as argument to >> tpm2_pcr_read(). The check is performed after information about the PCR >> banks has been retrieved. >> >> Signed-off-by: Roberto Sassu <roberto.sassu@huawei.com> > > What is the scenarion when this can happen (should be explained in > the commit message)? Without an HMAC session, the request/response payload can be modified. This patch ensures that the digest size in the payload is equal to the size of the algorithm specified by the caller. Patch 3/5 only ensures that there is no buffer overflow when data is copied to the tpm_digest structure passed by the caller. Patch 5/5 uses the PCR bank information introduced in patch 4/5 to ensure that the exact amount of data is copied from the response payload. However, the patch may not help because an attacker can modify the algorithm in the request payload so that the TPM returns a shorter digest. For me it is ok to remove this patch from the set. It was requested by Mimi. Roberto
On Wed, 31 Oct 2018, Roberto Sassu wrote: > On 10/30/2018 8:52 PM, Jarkko Sakkinen wrote: >> On Tue, 30 Oct 2018, Roberto Sassu wrote: >>> This patch ensures that the digest size returned by the TPM during a PCR >>> read matches the size of the algorithm passed as argument to >>> tpm2_pcr_read(). The check is performed after information about the PCR >>> banks has been retrieved. >>> >>> Signed-off-by: Roberto Sassu <roberto.sassu@huawei.com> >> >> What is the scenarion when this can happen (should be explained in >> the commit message)? > > Without an HMAC session, the request/response payload can be modified. > This patch ensures that the digest size in the payload is equal to the > size of the algorithm specified by the caller. i.e. it protect against memory corruption that could happen in the bus? Please state this. > For me it is ok to remove this patch from the set. It was requested by > Mimi. For me it is not ok remove this patch :-) I just want that note to the commit message in order to have it documented. /Jarkko
On Tue, 2018-10-30 at 16:47 +0100, Roberto Sassu wrote: > This patch ensures that the digest size returned by the TPM during a PCR > read matches the size of the algorithm passed as argument to > tpm2_pcr_read(). The check is performed after information about the PCR > banks has been retrieved. > > Signed-off-by: Roberto Sassu <roberto.sassu@huawei.com> > --- > drivers/char/tpm/tpm2-cmd.c | 16 +++++++++++++++- > 1 file changed, 15 insertions(+), 1 deletion(-) > > diff --git a/drivers/char/tpm/tpm2-cmd.c b/drivers/char/tpm/tpm2-cmd.c > index 8e821e7b4674..477dcc30fc53 100644 > --- a/drivers/char/tpm/tpm2-cmd.c > +++ b/drivers/char/tpm/tpm2-cmd.c > @@ -187,15 +187,28 @@ struct tpm2_pcr_read_out { > int tpm2_pcr_read(struct tpm_chip *chip, int pcr_idx, > struct tpm_digest *digest_struct, u16 *digest_size_ptr) > { > + int i; > int rc; > struct tpm_buf buf; > struct tpm2_pcr_read_out *out; > u8 pcr_select[TPM2_PCR_SELECT_MIN] = {0}; > u16 digest_size; > + u16 expected_digest_size = 0; > > if (pcr_idx >= TPM2_PLATFORM_PCR) > return -EINVAL; > > + if (!digest_size_ptr) { > + for (i = 0; i < ARRAY_SIZE(chip->active_banks) && > + chip->active_banks[i].alg_id != digest_struct->alg_id; i++) > + ; > + > + if (i == ARRAY_SIZE(chip->active_banks)) > + return -EINVAL; > + > + expected_digest_size = chip->active_banks[i].digest_size; > + } > + > rc = tpm_buf_init(&buf, TPM2_ST_NO_SESSIONS, TPM2_CC_PCR_READ); > if (rc) > return rc; > @@ -215,7 +228,8 @@ int tpm2_pcr_read(struct tpm_chip *chip, int pcr_idx, > > out = (struct tpm2_pcr_read_out *)&buf.data[TPM_HEADER_SIZE]; > digest_size = be16_to_cpu(out->digest_size); > - if (digest_size > sizeof(digest_struct->digest)) { > + if ((digest_size_ptr && digest_size > sizeof(digest_struct->digest)) || The returned digest size should never be larger than the structure field. The digest_size_ptr test is unnecessary. Mimi > + (!digest_size_ptr && digest_size != expected_digest_size)) { > rc = -EINVAL; > goto out; > }
diff --git a/drivers/char/tpm/tpm2-cmd.c b/drivers/char/tpm/tpm2-cmd.c index 8e821e7b4674..477dcc30fc53 100644 --- a/drivers/char/tpm/tpm2-cmd.c +++ b/drivers/char/tpm/tpm2-cmd.c @@ -187,15 +187,28 @@ struct tpm2_pcr_read_out { int tpm2_pcr_read(struct tpm_chip *chip, int pcr_idx, struct tpm_digest *digest_struct, u16 *digest_size_ptr) { + int i; int rc; struct tpm_buf buf; struct tpm2_pcr_read_out *out; u8 pcr_select[TPM2_PCR_SELECT_MIN] = {0}; u16 digest_size; + u16 expected_digest_size = 0; if (pcr_idx >= TPM2_PLATFORM_PCR) return -EINVAL; + if (!digest_size_ptr) { + for (i = 0; i < ARRAY_SIZE(chip->active_banks) && + chip->active_banks[i].alg_id != digest_struct->alg_id; i++) + ; + + if (i == ARRAY_SIZE(chip->active_banks)) + return -EINVAL; + + expected_digest_size = chip->active_banks[i].digest_size; + } + rc = tpm_buf_init(&buf, TPM2_ST_NO_SESSIONS, TPM2_CC_PCR_READ); if (rc) return rc; @@ -215,7 +228,8 @@ int tpm2_pcr_read(struct tpm_chip *chip, int pcr_idx, out = (struct tpm2_pcr_read_out *)&buf.data[TPM_HEADER_SIZE]; digest_size = be16_to_cpu(out->digest_size); - if (digest_size > sizeof(digest_struct->digest)) { + if ((digest_size_ptr && digest_size > sizeof(digest_struct->digest)) || + (!digest_size_ptr && digest_size != expected_digest_size)) { rc = -EINVAL; goto out; }
This patch ensures that the digest size returned by the TPM during a PCR read matches the size of the algorithm passed as argument to tpm2_pcr_read(). The check is performed after information about the PCR banks has been retrieved. Signed-off-by: Roberto Sassu <roberto.sassu@huawei.com> --- drivers/char/tpm/tpm2-cmd.c | 16 +++++++++++++++- 1 file changed, 15 insertions(+), 1 deletion(-)