From: Vitaly Chikunov <vt@altlinux.org>
To: Tianjia Zhang <tianjia.zhang@linux.alibaba.com>
Cc: Mimi Zohar <zohar@linux.ibm.com>,
linux-integrity@vger.kernel.org,
Jia Zhang <zhang.jia@linux.alibaba.com>
Subject: Re: [PATCH ima-evm-utils] ima-evm-utils: ima_sign supports SM2 and SM3 algorithm combination
Date: Sun, 31 Jan 2021 21:52:25 +0300 [thread overview]
Message-ID: <20210131185225.jemxbsyweq4vm7ex@altlinux.org> (raw)
In-Reply-To: <20210131175623.pbtdmeyvyeoenk3j@altlinux.org>
Tianjia,
On Sun, Jan 31, 2021 at 08:56:23PM +0300, Vitaly Chikunov wrote:
> On Sun, Jan 31, 2021 at 08:48:46PM +0300, Vitaly Chikunov wrote:
> > On Sun, Jan 31, 2021 at 11:27:21AM +0800, Tianjia Zhang wrote:
> > > The combination of SM2 and SM3 algorithms has been implemented in the
> > > kernel. At present, the ima-evm-utils signature tool does not support
> > > this combination of algorithms. This is because SM2 sign require a
> > > USERID, which requires the use of a higher-level sign functions of
> > > OpenSSL. this patch use the EVP_DigestSign series of functions to
> > > sign to support various signature algorithm combinations.
> > >
> > > Signed-off-by: Tianjia Zhang <tianjia.zhang@linux.alibaba.com>
> > > ---
> > > src/libimaevm.c | 37 +++++++++++++++++++++++++++++++------
> > > 1 file changed, 31 insertions(+), 6 deletions(-)
> > >
> > > diff --git a/src/libimaevm.c b/src/libimaevm.c
> > > index fa6c278..89b9b88 100644
> > > --- a/src/libimaevm.c
> > > +++ b/src/libimaevm.c
> > > @@ -891,6 +891,7 @@ static int sign_hash_v2(const char *algo, const unsigned char *hash,
> > > EVP_PKEY *pkey;
> > > char name[20];
> > > EVP_PKEY_CTX *ctx = NULL;
> > > + EVP_MD_CTX *mctx = NULL;
> > > const EVP_MD *md;
> > > size_t sigsize;
> > > const char *st;
> > > @@ -932,24 +933,47 @@ static int sign_hash_v2(const char *algo, const unsigned char *hash,
> > > return -1;
> > > }
> > >
> > > +#if OPENSSL_VERSION_NUMBER < 0x30000000
> >
> > Perhaps, this check is not needed at all if it isn't enabled for new
> > openssl?
> >
> >
> > > + /*
> > > + * SM2 and SM3 should go together. If SM3 hash algorithm and EC private
> > > + * key are used at the same time, check whether it is SM2 private key.
> > > + */
> > > + if (hdr->hash_algo == PKEY_HASH_SM3_256 && EVP_PKEY_id(pkey) == EVP_PKEY_EC) {
> > > + EC_KEY *ec = EVP_PKEY_get0_EC_KEY(pkey);
> > > + int curve = EC_GROUP_get_curve_name(EC_KEY_get0_group(ec));
> > > + if (curve == NID_sm2)
Also, this patch does not compile on openssl 1.1.0 because
NID_sm2 and EVP_PKEY_SM2 are not defined.
> > > + EVP_PKEY_set_alias_type(pkey, EVP_PKEY_SM2);
> > > + }
> > > +#endif
> > > +
> > > calc_keyid_v2(&keyid, name, pkey);
> > > hdr->keyid = keyid;
> > >
> > > st = "EVP_PKEY_CTX_new";
> > > if (!(ctx = EVP_PKEY_CTX_new(pkey, NULL)))
> > > goto err;
> > > - st = "EVP_PKEY_sign_init";
> > > - if (!EVP_PKEY_sign_init(ctx))
> > > + st = "EVP_MD_CTX_new";
> > > + if (!(mctx = EVP_MD_CTX_new()))
> > > goto err;
> > > + if (EVP_PKEY_id(pkey) == EVP_PKEY_SM2) {
> > > + st = "EVP_PKEY_CTX_set1_id";
> > > + /* Set SM2 default userid */
> > > + if (!EVP_PKEY_CTX_set1_id(ctx, "1234567812345678", 16))
> >
> > You cannot call EVP_PKEY_CTX_set1_id for EVP_PKEY_sign's ctx?
> > I don't really get the API change. Can you explain more this
> > requirement?
>
> EVP_PKEY_sign should be able to sign with any algo, unless there is
> openssl bug. Is there bug preventing SM2 signing with EVP_PKEY_sign?
>
> > > + goto err;
> > > + }
> > > + EVP_MD_CTX_set_pkey_ctx(mctx, ctx);
> > > st = "EVP_get_digestbyname";
> > > if (!(md = EVP_get_digestbyname(imaevm_params.hash_algo)))
> > > goto err;
> > > - st = "EVP_PKEY_CTX_set_signature_md";
> > > - if (!EVP_PKEY_CTX_set_signature_md(ctx, md))
> > > + st = "EVP_DigestSignInit";
> > > + if (!EVP_DigestSignInit(mctx, NULL, md, NULL, pkey))
> > > + goto err;
> > > + st = "EVP_DigestSignUpdate";
> > > + if (!EVP_DigestSignUpdate(mctx, hash, size))
> > > goto err;
> > > - st = "EVP_PKEY_sign";
> > > + st = "EVP_DigestSignFinal";
> > > sigsize = MAX_SIGNATURE_SIZE - sizeof(struct signature_v2_hdr) - 1;
> > > - if (!EVP_PKEY_sign(ctx, hdr->sig, &sigsize, hash, size))
Also, this conversion to EVP_DigestSign is incorrect, because
EVP_PKEY_sign does not hash the data (and expected to work over a ready
digest, see man EVP_PKEY_sign). While EVP_DigestSignFinal does hash the
data.
You should really try to run `make check` before sending patches.
Thanks,
> > > + if (!EVP_DigestSignFinal(mctx, hdr->sig, &sigsize))
> > > goto err;
> > > len = (int)sigsize;
> > >
> > > @@ -964,6 +988,7 @@ err:
> > > ERR_reason_error_string(ERR_peek_error()), st);
> > > output_openssl_errors();
> > > }
> > > + EVP_MD_CTX_free(mctx);
> > > EVP_PKEY_CTX_free(ctx);
> > > EVP_PKEY_free(pkey);
> > > return len;
> > > --
> > > 2.19.1.3.ge56e4f7
next prev parent reply other threads:[~2021-01-31 19:37 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-01-31 3:27 [PATCH ima-evm-utils] ima-evm-utils: ima_sign supports SM2 and SM3 algorithm combination Tianjia Zhang
2021-01-31 17:48 ` Vitaly Chikunov
2021-01-31 17:56 ` Vitaly Chikunov
2021-01-31 18:52 ` Vitaly Chikunov [this message]
2021-02-01 9:10 ` Tianjia Zhang
2021-02-01 12:31 ` Vitaly Chikunov
2021-02-11 5:39 ` Tianjia Zhang
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=20210131185225.jemxbsyweq4vm7ex@altlinux.org \
--to=vt@altlinux.org \
--cc=linux-integrity@vger.kernel.org \
--cc=tianjia.zhang@linux.alibaba.com \
--cc=zhang.jia@linux.alibaba.com \
--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
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox