Linux-Integrity Archive on lore.kernel.org
 help / color / Atom feed
From: James Bottomley <James.Bottomley@HansenPartnership.com>
To: David Woodhouse <dwmw2@infradead.org>,
	linux-integrity@vger.kernel.org,
	David Howells <dhowells@redhat.com>
Cc: Mimi Zohar <zohar@linux.ibm.com>,
	Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
Subject: Re: [PATCH 4/8] security: keys: trusted: use ASN.1 tpm2 key format for the blobs
Date: Mon, 09 Dec 2019 08:31:59 -0800
Message-ID: <1575909119.3340.27.camel@HansenPartnership.com> (raw)
In-Reply-To: <3b035027b934eb253143dca66ebae4356b386efe.camel@infradead.org>

On Mon, 2019-12-09 at 10:04 +0000, David Woodhouse wrote:
> On Sat, 2019-12-07 at 21:10 -0800, James Bottomley wrote:
> > Modify the tpm2 key format blob output to export and import in the
> > ASN.1 form for tpm2 sealed object keys.  For compatibility with
> > prior trusted keys, the importer will also accept two tpm2b
> > quantities representing the public and private parts of the
> > key.  However, the export via keyctl pipe will only output the
> > ASN.1 format.
> 
> You still have a tpm2_key_encode() function which spits out the raw
> private/public blobs each prefixed with a length word. What's that
> still used for?

It's the TPM2B format for input of the keys to the TPM.

> > The benefit of the ASN.1 format is that it's a standard
> 
> We should probably make that true. Did we even get as far as writing
> up an RFC-style description of the ASN.1? 

No ... I was going to ask you if you'd made a start.  I take it linux-
integrity is as good a list as any to begin this on.  I'll cc the
openssl_tpm2_engine list ... are there any others?

> >  and thus the
> > exported key can be used by userspace tools.  The format includes
> > policy specifications, thus it gets us out of having to construct
> > policy handles in userspace and the format includes the parent
> > meaning you don't have to keep passing it in each time.
> > 
> > This patch only implements basic handling for the ASN.1 format, so
> > keys with passwords but no policy.
> 
> ... but doesn't bail out with an error when it sees something it
> doesn't yet understand? Including the 'secret' field which is only
> relevant for importable keys, etc.

I hadn't actually got around to defining importable blobs ... but I
think keeping the same OID and adding the shared import secret in [2]
would be good enough?  In which case the TPM will error out for us
because it won't be able to unwrap the private blob.  If keyctl were
capable of returning information about the source of the problem it
might be worth checking, but while it can't, I think getting -EINVAL
from the TPM is as good as putting a check returning -EINVAL.

> > Signed-off-by: James Bottomley
> > <James.Bottomley@HansenPartnership.com>
> > ---
> >  security/keys/trusted-keys/Makefile       |   2 +-
> >  security/keys/trusted-keys/tpm2key.asn1   |  23 ++++
> >  security/keys/trusted-keys/trusted_tpm1.c |   2 +-
> >  security/keys/trusted-keys/trusted_tpm2.c | 170
> > +++++++++++++++++++++++++++++-
> >  4 files changed, 190 insertions(+), 7 deletions(-)
> >  create mode 100644 security/keys/trusted-keys/tpm2key.asn1
> > 
> > diff --git a/security/keys/trusted-keys/Makefile
> > b/security/keys/trusted-keys/Makefile
> > index 7b73cebbb378..e0198641eff2 100644
> > --- a/security/keys/trusted-keys/Makefile
> > +++ b/security/keys/trusted-keys/Makefile
> > @@ -5,4 +5,4 @@
> >  
> >  obj-$(CONFIG_TRUSTED_KEYS) += trusted.o
> >  trusted-y += trusted_tpm1.o
> > -trusted-y += trusted_tpm2.o
> > +trusted-y += trusted_tpm2.o tpm2key.asn1.o
> > diff --git a/security/keys/trusted-keys/tpm2key.asn1
> > b/security/keys/trusted-keys/tpm2key.asn1
> > new file mode 100644
> > index 000000000000..1851b7c80f08
> > --- /dev/null
> > +++ b/security/keys/trusted-keys/tpm2key.asn1
> > @@ -0,0 +1,23 @@
> > +---
> > +--- Note: This isn't quite the definition in the standard
> > +---       However, the Linux asn.1 parser doesn't understand
> > +---       [2] EXPLICIT SEQUENCE OF OPTIONAL
> > +---       So there's an extra intermediate TPMPolicySequence
> > +---       definition to work around this
> 
> At the very least we should prod David with a pointy stick on that
> topic, rather than quietly working around it.

OK, I'll add him to the cc list on this one.

[...]
> > +static int tpm2_key_encode(struct trusted_key_payload *payload,
> > +			   struct trusted_key_options *options,
> > +			   u8 *src, u32 len)
> > +{
> > +	u8 *scratch = kmalloc(PAGE_SIZE, GFP_KERNEL);
> > +	u8 *work = scratch, *work1;
> > +	u8 *priv, *pub;
> > +	u16 priv_len, pub_len;
> > +
> > +	priv_len = get_unaligned_be16(src);
> > +	src += 2;
> > +	priv = src;
> > +	src += priv_len;
> > +	pub_len = get_unaligned_be16(src);
> > +	src += 2;
> > +	pub = src;
> > +
> > +	if (!scratch)
> > +		return -ENOMEM;
> > +
> > +	asn1_encode_oid(&work, tpm2key_oid,
> > asn1_oid_len(tpm2key_oid));
> > +	if (options->blobauth[0] == 0) {
> > +		unsigned char bool[3], *w = bool;
> > +		/* tag 0 is emptyAuth */
> > +		asn1_encode_boolean(&w, true);
> > +		asn1_encode_tag(&work, 0, bool, w - bool);
> > +	}
> > +	asn1_encode_integer(&work, options->keyhandle);
> > +	asn1_encode_octet_string(&work, pub, pub_len);
> > +	asn1_encode_octet_string(&work, priv, priv_len);
> > +
> > +	work1 = payload->blob;
> > +	asn1_encode_sequence(&work1, scratch, work - scratch);
> > +
> > +	return work1 - payload->blob;
> > +}
> 
> I still don't like the lack of overflow protection here, one layer up
> from the underlying encoding APIs I already commented on.

Fixed.

> > +struct tpm2key_context {
> > +	u32 parent;
> > +	const u8 *pub;
> > +	u32 pub_len;
> > +	const u8 *priv;
> > +	u32 priv_len;
> > +};
> > +
> > +static int tpm2_key_decode(struct trusted_key_payload *payload,
> > +			   struct trusted_key_options *options,
> > +			   u8 **buf)
> > +{
> > +	int ret;
> > +	struct tpm2key_context ctx;
> > +	u8 *blob;
> > +
> > +	ret = asn1_ber_decoder(&tpm2key_decoder, &ctx, payload-
> > >blob,
> > +			       payload->blob_len);
> > +	if (ret < 0)
> > +		return ret;
> > +
> > +	if (ctx.priv_len + ctx.pub_len > MAX_BLOB_SIZE)
> > +		return -EINVAL;
> > +
> > +	blob = kmalloc(ctx.priv_len + ctx.pub_len + 4,
> > GFP_KERNEL);
> > +	if (!blob)
> > +		return -ENOMEM;
> > +
> > +	*buf = blob;
> > +	options->keyhandle = ctx.parent;
> > +	put_unaligned_be16(ctx.priv_len, blob);
> > +	blob += 2;
> > +	memcpy(blob, ctx.priv, ctx.priv_len);
> > +	blob += ctx.priv_len;
> > +	put_unaligned_be16(ctx.pub_len, blob);
> > +	blob += 2;
> > +	memcpy(blob, ctx.pub, ctx.pub_len);
> > 
> 
> Hm, do we really have to create this legacy form here and pass it
> around? Can't we change whatever consumes this?

It's not a legacy form ... it's the intrinsic TPM2B form the TPM 2.0
uses so it's the natural destination form for the conversion.

James

  reply index

Thread overview: 32+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-12-08  5:06 [PATCH 0/8] Fix TPM 2.0 trusted keys James Bottomley
2019-12-08  5:07 ` [PATCH 1/8] security: keys: trusted: flush the key handle after use James Bottomley
2019-12-09  8:31   ` David Woodhouse
2019-12-09 15:38     ` James Bottomley
2019-12-08  5:08 ` [PATCH 2/8] lib: add asn.1 encoder James Bottomley
2019-12-09  8:50   ` David Woodhouse
2019-12-09 15:46     ` James Bottomley
2019-12-09 22:05   ` Matthew Garrett
2019-12-09 22:43     ` James Bottomley
2019-12-08  5:09 ` [PATCH 3/8] oid_registry: Add TCG defined OIDS for TPM keys James Bottomley
2019-12-09  8:55   ` David Woodhouse
2019-12-09 16:21     ` James Bottomley
2020-06-19 20:45     ` Wiseman, Monty (GE Research, US)
2020-06-19 22:50       ` Jerry Snitselaar
2020-06-20 15:36       ` James Bottomley
2020-06-23  1:17       ` Jarkko Sakkinen
2019-12-08  5:10 ` [PATCH 4/8] security: keys: trusted: use ASN.1 tpm2 key format for the blobs James Bottomley
2019-12-09 10:04   ` David Woodhouse
2019-12-09 16:31     ` James Bottomley [this message]
2019-12-08  5:11 ` [PATCH 5/8] security: keys: trusted: Make sealed key properly interoperable James Bottomley
2019-12-09 10:09   ` David Woodhouse
2019-12-09 17:23     ` James Bottomley
2019-12-08  5:12 ` [PATCH 6/8] security: keys: trusted: add PCR policy to TPM2 keys James Bottomley
2019-12-09 10:18   ` David Woodhouse
2019-12-09 18:03     ` James Bottomley
2019-12-09 18:44       ` David Woodhouse
2019-12-09 19:11         ` James Bottomley
2019-12-25 17:08           ` Ken Goldman
2019-12-08  5:13 ` [PATCH 7/8] security: keys: trusted: add ability to specify arbitrary policy James Bottomley
2019-12-08  5:14 ` [PATCH 8/8] security: keys: trusted: implement counter/timer policy James Bottomley
2019-12-09 20:20 ` [PATCH 0/8] Fix TPM 2.0 trusted keys Jarkko Sakkinen
2019-12-09 20:57   ` James Bottomley

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1575909119.3340.27.camel@HansenPartnership.com \
    --to=james.bottomley@hansenpartnership.com \
    --cc=dhowells@redhat.com \
    --cc=dwmw2@infradead.org \
    --cc=jarkko.sakkinen@linux.intel.com \
    --cc=linux-integrity@vger.kernel.org \
    --cc=zohar@linux.ibm.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

Linux-Integrity Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/linux-integrity/0 linux-integrity/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 linux-integrity linux-integrity/ https://lore.kernel.org/linux-integrity \
		linux-integrity@vger.kernel.org
	public-inbox-index linux-integrity

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.kernel.vger.linux-integrity


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git