2021-07-14 09:30:11

by Ahmad Fatoum

[permalink] [raw]
Subject: Re: [PATCH 2/3] KEYS: trusted: Introduce support for NXP DCP-based trusted keys

Hello Richard,

On 14.06.21 22:16, Richard Weinberger wrote:
> DCP (Data Co-Processor) is the little brother of NXP's CAAM IP.
>
> Beside of accelerated crypto operations, it also offers support for
> hardware-bound keys. Using this feature it is possible to implement a blob
> mechanism just like CAAM offers. Unlike on CAAM, constructing and
> parsing the blob has to happen in software.
>
> We chose the following format for the blob:
> /*
> * struct dcp_blob_fmt - DCP BLOB format.
> *
> * @fmt_version: Format version, currently being %1
> * @blob_key: Random AES 128 key which is used to encrypt @payload,
> * @blob_key itself is encrypted with OTP or UNIQUE device key in
> * AES-128-ECB mode by DCP.
> * @nonce: Random nonce used for @payload encryption.
> * @payload_len: Length of the plain text @payload.
> * @payload: The payload itself, encrypted using AES-128-GCM and @blob_key,
> * GCM auth tag of size AES_BLOCK_SIZE is attached at the end of it.
> *
> * The total size of a DCP BLOB is sizeof(struct dcp_blob_fmt) + @payload_len +
> * AES_BLOCK_SIZE.
> */
> struct dcp_blob_fmt {
> __u8 fmt_version;
> __u8 blob_key[AES_KEYSIZE_128];
> __u8 nonce[AES_KEYSIZE_128];
> __le32 payload_len;
> __u8 payload[0];

There's been ongoing effort to replace the [0] GNU extension with C99
flexible array members. Please use [] here as well.

> } __packed;
>
> @payload is the key provided by trusted_key_ops->seal().
>
> By default the UNIQUE device key is used, it is also possible to use
> the OTP key. While the UNIQUE device key should be unique it is not
> entirely clear whether this is the case due to unclear documentation.
> If someone wants to be sure they can burn their own unique key
> into the OTP fuse and set the use_otp_key module parameter.
>
> Cc: Ahmad Fatoum <[email protected]>
> Cc: David Gstir <[email protected]>
> Cc: David Howells <[email protected]>
> Cc: "David S. Miller" <[email protected]>
> Cc: Fabio Estevam <[email protected]>
> Cc: Herbert Xu <[email protected]>
> Cc: James Bottomley <[email protected]>
> Cc: James Morris <[email protected]>
> Cc: Jarkko Sakkinen <[email protected]>
> Cc: Jonathan Corbet <[email protected]>
> Cc: [email protected]
> Cc: [email protected]
> Cc: [email protected]
> Cc: [email protected]
> Cc: [email protected]
> Cc: [email protected]
> Cc: [email protected]
> Cc: Mimi Zohar <[email protected]>
> Cc: NXP Linux Team <[email protected]>
> Cc: Pengutronix Kernel Team <[email protected]>
> Cc: Richard Weinberger <[email protected]>
> Cc: Sascha Hauer <[email protected]>
> Cc: "Serge E. Hallyn" <[email protected]>
> Cc: Shawn Guo <[email protected]>
> Co-developed-by: David Gstir <[email protected]>
> Signed-off-by: David Gstir <[email protected]>
> Signed-off-by: Richard Weinberger <[email protected]>
> ---
> .../admin-guide/kernel-parameters.txt | 1 +
> MAINTAINERS | 9 +
> include/keys/trusted_dcp.h | 13 +
> security/keys/trusted-keys/Makefile | 1 +
> security/keys/trusted-keys/trusted_core.c | 6 +-
> security/keys/trusted-keys/trusted_dcp.c | 325 ++++++++++++++++++
> 6 files changed, 354 insertions(+), 1 deletion(-)
> create mode 100644 include/keys/trusted_dcp.h
> create mode 100644 security/keys/trusted-keys/trusted_dcp.c
>
> diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
> index cb89dbdedc46..d8b2aa94a566 100644
> --- a/Documentation/admin-guide/kernel-parameters.txt
> +++ b/Documentation/admin-guide/kernel-parameters.txt
> @@ -5639,6 +5639,7 @@
> sources:
> - "tpm"
> - "tee"
> + - "dcp"
> If not specified then it defaults to iterating through
> the trust source list starting with TPM and assigns the
> first trust source as a backend which is initialized
> diff --git a/MAINTAINERS b/MAINTAINERS
> index b706dd20ff2b..779cac00827c 100644
> --- a/MAINTAINERS
> +++ b/MAINTAINERS
> @@ -10099,6 +10099,15 @@ F: include/keys/trusted-type.h
> F: include/keys/trusted_tpm.h
> F: security/keys/trusted-keys/
>
> +KEYS-TRUSTED-DCP
> +M: David Gstir <[email protected]>
> +M: Richard Weinberger <[email protected]>
> +L: [email protected]
> +L: [email protected]
> +S: Supported
> +F: include/keys/trusted_dcp.h
> +F: security/keys/trusted-keys/trusted_dcp.c

Hmm, I didn't add a MAINTAINERS entry for CAAM trusted keys. Do you think I should?

> +
> KEYS-TRUSTED-TEE
> M: Sumit Garg <[email protected]>
> L: [email protected]
> diff --git a/include/keys/trusted_dcp.h b/include/keys/trusted_dcp.h
> new file mode 100644
> index 000000000000..7b2a1275c527
> --- /dev/null
> +++ b/include/keys/trusted_dcp.h
> @@ -0,0 +1,13 @@
> +/* SPDX-License-Identifier: GPL-2.0-only */
> +/*
> + * Copyright (C) 2021 sigma star gmbh
> + * Authors: David Gstir <[email protected]>
> + * Richard Weinberger <[email protected]>
> + */
> +
> +#ifndef TRUSTED_DCP_H
> +#define TRUSTED_DCP_H
> +
> +extern struct trusted_key_ops dcp_trusted_key_ops;
> +
> +#endif
> diff --git a/security/keys/trusted-keys/Makefile b/security/keys/trusted-keys/Makefile
> index feb8b6c3cc79..992b591692dc 100644
> --- a/security/keys/trusted-keys/Makefile
> +++ b/security/keys/trusted-keys/Makefile
> @@ -12,3 +12,4 @@ trusted-y += trusted_tpm2.o
> trusted-y += tpm2key.asn1.o
>
> trusted-$(CONFIG_TEE) += trusted_tee.o
> +trusted-$(CONFIG_CRYPTO_DEV_MXS_DCP) += trusted_dcp.o
> diff --git a/security/keys/trusted-keys/trusted_core.c b/security/keys/trusted-keys/trusted_core.c
> index d5c891d8d353..66b631ffe876 100644
> --- a/security/keys/trusted-keys/trusted_core.c
> +++ b/security/keys/trusted-keys/trusted_core.c
> @@ -8,6 +8,7 @@
>
> #include <keys/user-type.h>
> #include <keys/trusted-type.h>
> +#include <keys/trusted_dcp.h>
> #include <keys/trusted_tee.h>
> #include <keys/trusted_tpm.h>
> #include <linux/capability.h>
> @@ -24,7 +25,7 @@
>
> static char *trusted_key_source;
> module_param_named(source, trusted_key_source, charp, 0);
> -MODULE_PARM_DESC(source, "Select trusted keys source (tpm or tee)");
> +MODULE_PARM_DESC(source, "Select trusted keys source (tpm, tee or dcp)");
>
> static const struct trusted_key_source trusted_key_sources[] = {
> #if defined(CONFIG_TCG_TPM)
> @@ -33,6 +34,9 @@ static const struct trusted_key_source trusted_key_sources[] = {
> #if defined(CONFIG_TEE)
> { "tee", &trusted_key_tee_ops },
> #endif
> +#if defined(CONFIG_CRYPTO_DEV_MXS_DCP)
> + { "dcp", &dcp_trusted_key_ops },
> +#endif
> };
>
> DEFINE_STATIC_CALL_NULL(trusted_key_init, *trusted_key_sources[0].ops->init);
> diff --git a/security/keys/trusted-keys/trusted_dcp.c b/security/keys/trusted-keys/trusted_dcp.c
> new file mode 100644
> index 000000000000..02db5dc261c7
> --- /dev/null
> +++ b/security/keys/trusted-keys/trusted_dcp.c
> @@ -0,0 +1,325 @@
> +// SPDX-License-Identifier: GPL-2.0-only
> +/*
> + * Copyright (C) 2021 sigma star gmbh
> + * Authors: David Gstir <[email protected]>
> + * Richard Weinberger <[email protected]>
> + */
> +
> +#include <crypto/aead.h>
> +#include <crypto/aes.h>
> +#include <crypto/algapi.h>
> +#include <crypto/gcm.h>
> +#include <crypto/skcipher.h>
> +#include <keys/trusted-type.h>
> +#include <linux/key-type.h>
> +#include <linux/module.h>
> +#include <linux/mxs-dcp.h>
> +#include <linux/printk.h>
> +#include <linux/random.h>
> +#include <linux/scatterlist.h>
> +
> +#define DCP_BLOB_VERSION 1
> +#define DCP_BLOB_AUTHLEN 16
> +
> +/**
> + * struct dcp_blob_fmt - DCP BLOB format.
> + *
> + * @fmt_version: Format version, currently being %1.
> + * @blob_key: Random AES 128 key which is used to encrypt @payload,
> + * @blob_key itself is encrypted with OTP or UNIQUE device key in
> + * AES-128-ECB mode by DCP.
> + * @nonce: Random nonce used for @payload encryption.
> + * @payload_len: Length of the plain text @payload.
> + * @payload: The payload itself, encrypted using AES-128-GCM and @blob_key,
> + * GCM auth tag of size DCP_BLOB_AUTHLEN is attached at the end of it.
> + *
> + * The total size of a DCP BLOB is sizeof(struct dcp_blob_fmt) + @payload_len +
> + * DCP_BLOB_AUTHLEN.
> + */
> +struct dcp_blob_fmt {
> + __u8 fmt_version;
> + __u8 blob_key[AES_KEYSIZE_128];
> + __u8 nonce[AES_KEYSIZE_128];
> + __le32 payload_len;
> + __u8 payload[0];

[], see above.

> +} __packed;
> +
> +static bool use_otp_key;
> +module_param_named(dcp_use_otp_key, use_otp_key, bool, 0);
> +MODULE_PARM_DESC(dcp_use_otp_key, "Use OTP instead of UNIQUE key for sealing");

Shouldn't these be documented in admin-guide/kernel-parameters.txt as well?

> +static bool skip_zk_test;
> +module_param_named(dcp_skip_zk_test, skip_zk_test, bool, 0);
> +MODULE_PARM_DESC(dcp_skip_zk_test, "Don't test whether device keys are zero'ed");

Does this need to be configurible? I'd assume this can only happen when using an
unfused OTP. In such a case, it's ok to always warn, so you don't need to make
this configurible.

> +
> +static unsigned int calc_blob_len(unsigned int payload_len)
> +{
> + return sizeof(struct dcp_blob_fmt) + payload_len + DCP_BLOB_AUTHLEN;
> +}
> +
> +static int do_dcp_crypto(u8 *in, u8 *out, bool is_encrypt)

I assume in can't be const because the use with sg APIs?

> +{
> + int res = 0;
> + struct skcipher_request *req = NULL;
> + DECLARE_CRYPTO_WAIT(wait);
> + struct scatterlist src_sg, dst_sg;
> + struct crypto_skcipher *tfm;
> + u8 paes_key[DCP_PAES_KEYSIZE];
> +
> + if (!use_otp_key)

I'd invert this. Makes code easier to read.

> + paes_key[0] = DCP_PAES_KEY_UNIQUE;
> + else
> + paes_key[0] = DCP_PAES_KEY_OTP;
> +
> + tfm = crypto_alloc_skcipher("ecb-paes-dcp", CRYPTO_ALG_INTERNAL,
> + CRYPTO_ALG_INTERNAL);
> + if (IS_ERR(tfm)) {
> + res = PTR_ERR(tfm);
> + pr_err("Unable to request DCP pAES-ECB cipher: %i\n", res);

Can you define pr_fmt above? There's also %pe now that can directly print out an
error pointer.

> + tfm = NULL;
> + goto out;
> + }
> +
> + req = skcipher_request_alloc(tfm, GFP_NOFS);
> + if (!req) {
> + res = -ENOMEM;
> + goto out;
> + }
> +
> + skcipher_request_set_callback(req, CRYPTO_TFM_REQ_MAY_BACKLOG |
> + CRYPTO_TFM_REQ_MAY_SLEEP,
> + crypto_req_done, &wait);
> + res = crypto_skcipher_setkey(tfm, paes_key, sizeof(paes_key));
> + if (res < 0)
> + goto out;
> +
> + sg_init_one(&src_sg, in, AES_KEYSIZE_128);
> + sg_init_one(&dst_sg, out, AES_KEYSIZE_128);
> + skcipher_request_set_crypt(req, &src_sg, &dst_sg, AES_KEYSIZE_128,
> + NULL);
> +
> + if (is_encrypt)
> + res = crypto_wait_req(crypto_skcipher_encrypt(req), &wait);
> + else
> + res = crypto_wait_req(crypto_skcipher_decrypt(req), &wait);
> +
> +out:
> + skcipher_request_free(req);
> + crypto_free_skcipher(tfm);
> +
> + return res;
> +}
> +
> +static int do_aead_crypto(u8 *in, u8 *out, size_t len, u8 *key, u8 *nonce,
> + bool is_encrypt)
> +{
> + struct aead_request *aead_req = NULL;
> + struct scatterlist src_sg, dst_sg;
> + struct crypto_aead *aead;
> + int ret;
> +
> + aead = crypto_alloc_aead("gcm(aes)", 0, CRYPTO_ALG_ASYNC);
> + if (IS_ERR(aead)) {
> + ret = PTR_ERR(aead);
> + pr_err("Unable to request AES-GCM cipher: %i\n", ret);
> + goto out;
> + }
> +
> + ret = crypto_aead_setauthsize(aead, DCP_BLOB_AUTHLEN);
> + if (ret < 0) {
> + pr_err("Can't set crypto auth tag len: %d\n", ret);
> + goto free_aead;
> + }
> +
> + aead_req = aead_request_alloc(aead, GFP_KERNEL);
> + if (!aead_req) {
> + ret = -ENOMEM;
> + goto free_aead;
> + }
> +
> + sg_init_one(&src_sg, in, len);
> + if (is_encrypt) {
> + /*
> + * If we encrypt our buffer has extra space for the auth tag.
> + */
> + sg_init_one(&dst_sg, out, len + DCP_BLOB_AUTHLEN);
> + } else {
> + sg_init_one(&dst_sg, out, len);
> + }
> +
> + aead_request_set_crypt(aead_req, &src_sg, &dst_sg, len, nonce);
> + aead_request_set_callback(aead_req, CRYPTO_TFM_REQ_MAY_SLEEP, NULL,
> + NULL);
> + aead_request_set_ad(aead_req, 0);
> +
> + if (crypto_aead_setkey(aead, key, AES_KEYSIZE_128)) {
> + pr_err("Can't set crypto AEAD key\n");
> + ret = -EINVAL;
> + goto free_req;
> + }
> +
> + if (is_encrypt)
> + ret = crypto_aead_encrypt(aead_req);
> + else
> + ret = crypto_aead_decrypt(aead_req);
> +
> +free_req:
> + aead_request_free(aead_req);
> +free_aead:
> + crypto_free_aead(aead);
> +out:
> + return ret;
> +}
> +
> +static int decrypt_blob_key(u8 *key)
> +{
> + return do_dcp_crypto(key, key, false);
> +}
> +
> +static int encrypt_blob_key(u8 *key)
> +{
> + return do_dcp_crypto(key, key, true);
> +}
> +
> +static int trusted_dcp_seal(struct trusted_key_payload *p, char *datablob)
> +{
> + struct dcp_blob_fmt *b = (struct dcp_blob_fmt *)p->blob;
> + int blen, ret;
> +
> + blen = calc_blob_len(p->key_len);
> + if (blen > MAX_BLOB_SIZE) {
> + ret = -E2BIG;
> + goto out;

Nitpick, just return and drop the goto?

> + }
> +
> + b->fmt_version = DCP_BLOB_VERSION;
> + get_random_bytes(b->nonce, AES_KEYSIZE_128);
> + get_random_bytes(b->blob_key, AES_KEYSIZE_128);
> +
> + ret = do_aead_crypto(p->key, b->payload, p->key_len, b->blob_key,
> + b->nonce, true);
> + if (ret) {
> + pr_err("Unable to encrypt blob payload: %i\n", ret);
> + goto out;
> + }
> +
> + ret = encrypt_blob_key(b->blob_key);
> + if (ret) {
> + pr_err("Unable to encrypt blob key: %i\n", ret);
> + goto out;
> + }
> +
> + b->payload_len = cpu_to_le32(p->key_len);

- payload_len is at offset 33, but MIN_KEY_SIZE == 32 and there are no minimum
size checks. Couldn't you read beyond the buffer this way?

- offset 33 is unaligned for payload_len. Please use get_unaligned_le32 here.


> + p->blob_len = blen;
> + ret = 0;
> +out:
> + return ret;
> +}
> +
> +static int trusted_dcp_unseal(struct trusted_key_payload *p, char *datablob)
> +{
> + struct dcp_blob_fmt *b = (struct dcp_blob_fmt *)p->blob;
> + int blen, ret;
> +
> + if (b->fmt_version != DCP_BLOB_VERSION) {
> + pr_err("DCP blob has bad version: %i, expected %i\n",
> + b->fmt_version, DCP_BLOB_VERSION);
> + ret = -EINVAL;
> + goto out;
> + }
> +
> + p->key_len = le32_to_cpu(b->payload_len);
> + blen = calc_blob_len(p->key_len);
> + if (blen != p->blob_len) {
> + pr_err("DCP blob has bad length: %i != %i\n", blen,
> + p->blob_len);
> + ret = -EINVAL;
> + goto out;
> + }
> +
> + ret = decrypt_blob_key(b->blob_key);
> + if (ret) {
> + pr_err("Unable to decrypt blob key: %i\n", ret);
> + goto out;
> + }
> +
> + ret = do_aead_crypto(b->payload, p->key, p->key_len + DCP_BLOB_AUTHLEN,
> + b->blob_key, b->nonce, false);
> + if (ret) {
> + pr_err("Unwrap of DCP payload failed: %i\n", ret);
> + goto out;
> + }
> +
> + ret = 0;
> +out:
> + return ret;
> +}
> +
> +static int trusted_dcp_get_random(unsigned char *key, size_t key_len)
> +{
> + get_random_bytes(key, key_len);
> +
> + return key_len;
> +}

jfyi, in the prelude of my CAAM series, I made this the default
when .get_random == NULL.

> +
> +static int test_for_zero_key(void)
> +{
> + static const u8 bad[] = {0x9a, 0xda, 0xe0, 0x54, 0xf6, 0x3d, 0xfa, 0xff,
> + 0x5e, 0xa1, 0x8e, 0x45, 0xed, 0xf6, 0xea, 0x6f};
> + void *buf = NULL;
> + int ret = 0;
> +
> + if (skip_zk_test)
> + goto out;
> +
> + buf = kmalloc(AES_BLOCK_SIZE, GFP_KERNEL);
> + if (!buf) {
> + ret = -ENOMEM;
> + goto out;
> + }
> +
> + memset(buf, 0x55, AES_BLOCK_SIZE);
> +
> + ret = do_dcp_crypto(buf, buf, true);
> + if (ret)
> + goto out;
> +
> + if (memcmp(buf, bad, AES_BLOCK_SIZE) == 0) {
> + pr_err("Device neither in secure nor trusted mode!\n");

What's the difference between secure and trusted? Can't this test be skipped
if use_otp_key == false?

> + ret = -EINVAL;
> + }
> +out:
> + kfree(buf);
> + return ret;
> +}
> +
> +static int trusted_dcp_init(void)
> +{
> + int ret;
> +
> + if (use_otp_key)
> + pr_info("Using DCP OTP key\n");
> +
> + ret = test_for_zero_key();
> + if (ret) {
> + pr_err("Test for zero'ed keys failed: %i\n", ret);
> +
> + return -EINVAL;
> + }
> +
> + return register_key_type(&key_type_trusted);
> +}
> +
> +static void trusted_dcp_exit(void)
> +{
> + unregister_key_type(&key_type_trusted);
> +}
> +
> +struct trusted_key_ops dcp_trusted_key_ops = {
> + .exit = trusted_dcp_exit,
> + .init = trusted_dcp_init,
> + .seal = trusted_dcp_seal,
> + .unseal = trusted_dcp_unseal,
> + .get_random = trusted_dcp_get_random,
> + .migratable = 0,
> +};

Cheers,
Ahmad


--
Pengutronix e.K. | |
Steuerwalder Str. 21 | http://www.pengutronix.de/ |
31137 Hildesheim, Germany | Phone: +49-5121-206917-0 |
Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 |


2021-07-14 10:45:16

by Richard Weinberger

[permalink] [raw]
Subject: Re: [PATCH 2/3] KEYS: trusted: Introduce support for NXP DCP-based trusted keys

Ahmad,

----- Urspr√ľngliche Mail -----
> Von: "Ahmad Fatoum" <[email protected]>

[...]

>> /*
>> * struct dcp_blob_fmt - DCP BLOB format.
>> *
>> * @fmt_version: Format version, currently being %1
>> * @blob_key: Random AES 128 key which is used to encrypt @payload,
>> * @blob_key itself is encrypted with OTP or UNIQUE device key in
>> * AES-128-ECB mode by DCP.
>> * @nonce: Random nonce used for @payload encryption.
>> * @payload_len: Length of the plain text @payload.
>> * @payload: The payload itself, encrypted using AES-128-GCM and @blob_key,
>> * GCM auth tag of size AES_BLOCK_SIZE is attached at the end of it.
>> *
>> * The total size of a DCP BLOB is sizeof(struct dcp_blob_fmt) + @payload_len +
>> * AES_BLOCK_SIZE.
>> */
>> struct dcp_blob_fmt {
>> __u8 fmt_version;
>> __u8 blob_key[AES_KEYSIZE_128];
>> __u8 nonce[AES_KEYSIZE_128];
>> __le32 payload_len;
>> __u8 payload[0];
>
> There's been ongoing effort to replace the [0] GNU extension with C99
> flexible array members. Please use [] here as well.

Makes sense!

[...]

>> +KEYS-TRUSTED-DCP
>> +M: David Gstir <[email protected]>
>> +M: Richard Weinberger <[email protected]>
>> +L: [email protected]
>> +L: [email protected]
>> +S: Supported
>> +F: include/keys/trusted_dcp.h
>> +F: security/keys/trusted-keys/trusted_dcp.c
>
> Hmm, I didn't add a MAINTAINERS entry for CAAM trusted keys. Do you think I
> should?

Sure, why not? It shows that you will also in future take care of it.

[...]

>> +} __packed;
>> +
>> +static bool use_otp_key;
>> +module_param_named(dcp_use_otp_key, use_otp_key, bool, 0);
>> +MODULE_PARM_DESC(dcp_use_otp_key, "Use OTP instead of UNIQUE key for sealing");
>
> Shouldn't these be documented in admin-guide/kernel-parameters.txt as well?

Yes. Will do.

>> +static bool skip_zk_test;
>> +module_param_named(dcp_skip_zk_test, skip_zk_test, bool, 0);
>> +MODULE_PARM_DESC(dcp_skip_zk_test, "Don't test whether device keys are
>> zero'ed");
>
> Does this need to be configurible? I'd assume this can only happen when using an
> unfused OTP. In such a case, it's ok to always warn, so you don't need to make
> this configurible.

We found such a setting super useful while working with targets where the keys are
zero'ed for various reasons.
There are cases where you want to use/test trusted keys even when the master key
is void. Our detection logic does not only print a warning, it refuses to load
blobs. So IMHO the config knob makes sense.

>> +
>> +static unsigned int calc_blob_len(unsigned int payload_len)
>> +{
>> + return sizeof(struct dcp_blob_fmt) + payload_len + DCP_BLOB_AUTHLEN;
>> +}
>> +
>> +static int do_dcp_crypto(u8 *in, u8 *out, bool is_encrypt)
>
> I assume in can't be const because the use with sg APIs?

I'm pretty sure this was the main reason, but I can check again.

>> +{
>> + int res = 0;
>> + struct skcipher_request *req = NULL;
>> + DECLARE_CRYPTO_WAIT(wait);
>> + struct scatterlist src_sg, dst_sg;
>> + struct crypto_skcipher *tfm;
>> + u8 paes_key[DCP_PAES_KEYSIZE];
>> +
>> + if (!use_otp_key)
>
> I'd invert this. Makes code easier to read.

Ok. :-)

>> + paes_key[0] = DCP_PAES_KEY_UNIQUE;
>> + else
>> + paes_key[0] = DCP_PAES_KEY_OTP;
>> +
>> + tfm = crypto_alloc_skcipher("ecb-paes-dcp", CRYPTO_ALG_INTERNAL,
>> + CRYPTO_ALG_INTERNAL);
>> + if (IS_ERR(tfm)) {
>> + res = PTR_ERR(tfm);
>> + pr_err("Unable to request DCP pAES-ECB cipher: %i\n", res);
>
> Can you define pr_fmt above? There's also %pe now that can directly print out an
> error pointer.

pr_fmt is not defined on purpose. include/keys/trusted-type.h defines already one
and I assumed "trusted_key:" is the desired prefix for all kinds of trusted keys.

[...]

> - payload_len is at offset 33, but MIN_KEY_SIZE == 32 and there are no minimum
> size checks. Couldn't you read beyond the buffer this way?

The key has a minimum size of MIN_KEY_SIZE, but p->blob (being struct trusted_key_payload->blob[MAX_BLOB_SIZE])
is much larger.
So the assumption is that a DCP blob will always be smaller than MAX_BLOB_SIZE.

> - offset 33 is unaligned for payload_len. Please use get_unaligned_le32 here.

Oh yes. Makes sense!

[...]

>
> jfyi, in the prelude of my CAAM series, I made this the default
> when .get_random == NULL.

Right. :-)

[...]

>> + ret = do_dcp_crypto(buf, buf, true);
>> + if (ret)
>> + goto out;
>> +
>> + if (memcmp(buf, bad, AES_BLOCK_SIZE) == 0) {
>> + pr_err("Device neither in secure nor trusted mode!\n");
>
> What's the difference between secure and trusted? Can't this test be skipped
> if use_otp_key == false?

DCP has many modes of operation. Secure is one level above trusted.
For the gory details see "Security Reference Manual for the i.MX 6ULL Applications Processor".
I'm not sure whether all information my manual describes is publicly available so I
don't dare to copy&paste from it.

As David and I understood the logic, both OTP and UNIQUE keys can be zero'ed.
It is also possible that DCP has no support at all for these keys,
then you'll also get a zero key. That's why we have this check here.

Thanks,
//richard

2021-07-21 17:20:40

by Ahmad Fatoum

[permalink] [raw]
Subject: Re: [PATCH 2/3] KEYS: trusted: Introduce support for NXP DCP-based trusted keys

Hello Richard,

On 14.07.21 12:44, Richard Weinberger wrote:
> Ahmad,
>
> ----- Urspr√ľngliche Mail -----
>> Von: "Ahmad Fatoum" <[email protected]>
>
> [...]
>
> Sure, why not? It shows that you will also in future take care of it.

Good point. I did that for v3.

>
> [...]
>
>>> +} __packed;
>>> +
>>> +static bool use_otp_key;
>>> +module_param_named(dcp_use_otp_key, use_otp_key, bool, 0);
>>> +MODULE_PARM_DESC(dcp_use_otp_key, "Use OTP instead of UNIQUE key for sealing");
>>
>> Shouldn't these be documented in admin-guide/kernel-parameters.txt as well?
>
> Yes. Will do.
>
>>> +static bool skip_zk_test;
>>> +module_param_named(dcp_skip_zk_test, skip_zk_test, bool, 0);
>>> +MODULE_PARM_DESC(dcp_skip_zk_test, "Don't test whether device keys are
>>> zero'ed");
>>
>> Does this need to be configurible? I'd assume this can only happen when using an
>> unfused OTP. In such a case, it's ok to always warn, so you don't need to make
>> this configurible.
>
> We found such a setting super useful while working with targets where the keys are
> zero'ed for various reasons.
> There are cases where you want to use/test trusted keys even when the master key
> is void. Our detection logic does not only print a warning, it refuses to load
> blobs. So IMHO the config knob makes sense.

Ah, I missed that it refuses to continue in that case.

>
>>> +
>>> +static unsigned int calc_blob_len(unsigned int payload_len)
>>> +{
>>> + return sizeof(struct dcp_blob_fmt) + payload_len + DCP_BLOB_AUTHLEN;
>>> +}
>>> +
>>> +static int do_dcp_crypto(u8 *in, u8 *out, bool is_encrypt)
>>
>> I assume in can't be const because the use with sg APIs?
>
> I'm pretty sure this was the main reason, but I can check again.
>
>>> +{
>>> + int res = 0;
>>> + struct skcipher_request *req = NULL;
>>> + DECLARE_CRYPTO_WAIT(wait);
>>> + struct scatterlist src_sg, dst_sg;
>>> + struct crypto_skcipher *tfm;
>>> + u8 paes_key[DCP_PAES_KEYSIZE];
>>> +
>>> + if (!use_otp_key)
>>
>> I'd invert this. Makes code easier to read.
>
> Ok. :-)
>
>>> + paes_key[0] = DCP_PAES_KEY_UNIQUE;
>>> + else
>>> + paes_key[0] = DCP_PAES_KEY_OTP;
>>> +
>>> + tfm = crypto_alloc_skcipher("ecb-paes-dcp", CRYPTO_ALG_INTERNAL,
>>> + CRYPTO_ALG_INTERNAL);
>>> + if (IS_ERR(tfm)) {
>>> + res = PTR_ERR(tfm);
>>> + pr_err("Unable to request DCP pAES-ECB cipher: %i\n", res);
>>
>> Can you define pr_fmt above? There's also %pe now that can directly print out an
>> error pointer.
>
> pr_fmt is not defined on purpose. include/keys/trusted-type.h defines already one
> and I assumed "trusted_key:" is the desired prefix for all kinds of trusted keys.

Ah, all good then. I didn't define it for CAAM either, but forgot why I didn't
along the way. May've been the same reason.

> [...]
>
>> - payload_len is at offset 33, but MIN_KEY_SIZE == 32 and there are no minimum
>> size checks. Couldn't you read beyond the buffer this way?
>
> The key has a minimum size of MIN_KEY_SIZE, but p->blob (being struct trusted_key_payload->blob[MAX_BLOB_SIZE])
> is much larger.
> So the assumption is that a DCP blob will always be smaller than MAX_BLOB_SIZE.
>
>> - offset 33 is unaligned for payload_len. Please use get_unaligned_le32 here.
>
> Oh yes. Makes sense!
>
> [...]
>
>>
>> jfyi, in the prelude of my CAAM series, I made this the default
>> when .get_random == NULL.
>
> Right. :-)
>
> [...]
>
>>> + ret = do_dcp_crypto(buf, buf, true);
>>> + if (ret)
>>> + goto out;
>>> +
>>> + if (memcmp(buf, bad, AES_BLOCK_SIZE) == 0) {
>>> + pr_err("Device neither in secure nor trusted mode!\n");
>>
>> What's the difference between secure and trusted? Can't this test be skipped
>> if use_otp_key == false?
>
> DCP has many modes of operation. Secure is one level above trusted.
> For the gory details see "Security Reference Manual for the i.MX 6ULL Applications Processor".
> I'm not sure whether all information my manual describes is publicly available so I
> don't dare to copy&paste from it.
>
> As David and I understood the logic, both OTP and UNIQUE keys can be zero'ed.
> It is also possible that DCP has no support at all for these keys,
> then you'll also get a zero key. That's why we have this check here.

Thanks for the clarification.

Cheers,
Ahmad

>
> Thanks,
> //richard
>


--
Pengutronix e.K. | |
Steuerwalder Str. 21 | http://www.pengutronix.de/ |
31137 Hildesheim, Germany | Phone: +49-5121-206917-0 |
Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 |